an approach to cover more advertisers in adwords
TRANSCRIPT
![Page 1: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/1.jpg)
An Approach to Cover More Advertisers in Adwords
ABudhiraja and P Krishna Reddyamarbudhirajaresearchiiitacin
pkreddyiiitacin
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Introduction
Search engines have been deemed as the starting point of most of the web transactions
This results in a significantly larger user base making search engines an ideal avenue for businesses to reach potential consumers
In the meta-model of knowledge sharing of search engines advertising has become a major revenue earning source for their sustenance
According to IAB standards sponsored search is the most dominant form of online advertising covering almost 43 of the entire market
Introduction Sponsored Search
The model of search engine advertising is more popularly known as Adwords
When a user queries a search engine a list of search results and sponsored results or advertisements are displayed
Advertisers bid on search keywords and pay the search engine according to Pay Per Click (PPC) model to display the advertisement on the query page containing the desired keywords
Introduction Problem StatementSearch keywords follow a long tail frequency distribution with
a small and fat head of highly frequent keywords
a long but thin tail of less frequent keywords
During the keyword auctions there is a high competition for head keywords while there is little to no competition for the tail keywords
This leads to underutilization of ad space of a large number of tail keywords Also it leads to ignorance of a diverse set of potential consumers who could be captured by targeting tail keywords
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Background Model of AdwordsThe present model is considered as
the online bipartite graph matching with advertisers as one disjoint set and incoming queries are the other
When a new query comes in it is to be matched to a set of advertisers
The advertisers are then ranked and their ads are display in that ranked order
Background Adwords Architecture
1 Analyze Query
2 Retrieve Relevant Advertisers
3 Bidding
4 Ranking Advertisers
Background Coverage Patterns- Central Idea
The basic idea of Coverage Pattern is inspired from the set cover problem in set theory
Given a universe U and a family S of subsets of U a cover is a subfamily C S subof sets whose union is U
Using the same notion coverage patterns aim to identify items that cover certain percentage of the entire data
A keypoint to be mentioned is that coverage patterns aim at identifying that usually ldquodo notrdquo occur together in contrast to frequent patterns that identify patterns in data that occur together
Background Coverage Patterns - Notations
Let W be a set of webpages of a website W = w1 w2 hellip wN
Let D be a set of transactions from the click stream data D = T1 T2 hellip such that T Wsub
X is defined as a patterns of webpages such that X W X = subwp wq hellip
Twi denotes the set of transactions containing the webpage wi and its cardinality is denoted as |Twi|
Background Coverage Patterns - Definitions
Background Coverage Patterns - Definitions
Background Coverage Patterns
A pattern is interesting if it has a high CS and low OR
A high CS value indicates more number of visitors and a low OR value means less repetitions amongst the visitors
A pattern is said to be interesting if CS(X) gt minCS(X) OR(X) lt maxOR and RF(wi) gt minRF
Background Coverage Patterns - Example
Dataset
Assuming minRF - 02 minCS - 03 and maxOR- to be 05
Ta is 5 Tb is 7 and f Tf is 1 So RF for a is 05 for b is 07 and for f is 01
RF(f) = 01 lt 02 (minRF) f will be removed RF(a) = 05 gt 02and RF(b) = 07 gt 02 so a and b are not removed
ba is a candidate pattern (Order in a pattern is decreasing order of the RF)
The Coverage Set for ba is 12345678910 and |CSet ba| is 10 Hence CS = 1010 = 1 gt 03 (minCS)
The transactions containing ba together are 110 and Ta = 5 so the overlap ratio is 25 = 04 lt maxOR is a coverage pattern
Since pattern ba has CS gt minCS and OR gt maxOR and a and b satisfy the minRF requirements ba is a coverage pattern
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Proposed Approach Basic IdeaBecause of the nature of distribution of search keywords there is very less
competition for tail keywords As a result there are very less or no advertisers for such keywords
We noticed that if could combine such keywords into groups such that these keyword groups have a certain number of visitors we can utilize the ad space of such keywords
To perform the grouping of search keywords we employ the notion of query taxonomy to group semantically similar words
These groups are further mined from the logs in the form of coverage patterns
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 2: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/2.jpg)
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Introduction
Search engines have been deemed as the starting point of most of the web transactions
This results in a significantly larger user base making search engines an ideal avenue for businesses to reach potential consumers
In the meta-model of knowledge sharing of search engines advertising has become a major revenue earning source for their sustenance
According to IAB standards sponsored search is the most dominant form of online advertising covering almost 43 of the entire market
Introduction Sponsored Search
The model of search engine advertising is more popularly known as Adwords
When a user queries a search engine a list of search results and sponsored results or advertisements are displayed
Advertisers bid on search keywords and pay the search engine according to Pay Per Click (PPC) model to display the advertisement on the query page containing the desired keywords
Introduction Problem StatementSearch keywords follow a long tail frequency distribution with
a small and fat head of highly frequent keywords
a long but thin tail of less frequent keywords
During the keyword auctions there is a high competition for head keywords while there is little to no competition for the tail keywords
This leads to underutilization of ad space of a large number of tail keywords Also it leads to ignorance of a diverse set of potential consumers who could be captured by targeting tail keywords
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Background Model of AdwordsThe present model is considered as
the online bipartite graph matching with advertisers as one disjoint set and incoming queries are the other
When a new query comes in it is to be matched to a set of advertisers
The advertisers are then ranked and their ads are display in that ranked order
Background Adwords Architecture
1 Analyze Query
2 Retrieve Relevant Advertisers
3 Bidding
4 Ranking Advertisers
Background Coverage Patterns- Central Idea
The basic idea of Coverage Pattern is inspired from the set cover problem in set theory
Given a universe U and a family S of subsets of U a cover is a subfamily C S subof sets whose union is U
Using the same notion coverage patterns aim to identify items that cover certain percentage of the entire data
A keypoint to be mentioned is that coverage patterns aim at identifying that usually ldquodo notrdquo occur together in contrast to frequent patterns that identify patterns in data that occur together
Background Coverage Patterns - Notations
Let W be a set of webpages of a website W = w1 w2 hellip wN
Let D be a set of transactions from the click stream data D = T1 T2 hellip such that T Wsub
X is defined as a patterns of webpages such that X W X = subwp wq hellip
Twi denotes the set of transactions containing the webpage wi and its cardinality is denoted as |Twi|
Background Coverage Patterns - Definitions
Background Coverage Patterns - Definitions
Background Coverage Patterns
A pattern is interesting if it has a high CS and low OR
A high CS value indicates more number of visitors and a low OR value means less repetitions amongst the visitors
A pattern is said to be interesting if CS(X) gt minCS(X) OR(X) lt maxOR and RF(wi) gt minRF
Background Coverage Patterns - Example
Dataset
Assuming minRF - 02 minCS - 03 and maxOR- to be 05
Ta is 5 Tb is 7 and f Tf is 1 So RF for a is 05 for b is 07 and for f is 01
RF(f) = 01 lt 02 (minRF) f will be removed RF(a) = 05 gt 02and RF(b) = 07 gt 02 so a and b are not removed
ba is a candidate pattern (Order in a pattern is decreasing order of the RF)
The Coverage Set for ba is 12345678910 and |CSet ba| is 10 Hence CS = 1010 = 1 gt 03 (minCS)
The transactions containing ba together are 110 and Ta = 5 so the overlap ratio is 25 = 04 lt maxOR is a coverage pattern
Since pattern ba has CS gt minCS and OR gt maxOR and a and b satisfy the minRF requirements ba is a coverage pattern
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Proposed Approach Basic IdeaBecause of the nature of distribution of search keywords there is very less
competition for tail keywords As a result there are very less or no advertisers for such keywords
We noticed that if could combine such keywords into groups such that these keyword groups have a certain number of visitors we can utilize the ad space of such keywords
To perform the grouping of search keywords we employ the notion of query taxonomy to group semantically similar words
These groups are further mined from the logs in the form of coverage patterns
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 3: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/3.jpg)
Introduction
Search engines have been deemed as the starting point of most of the web transactions
This results in a significantly larger user base making search engines an ideal avenue for businesses to reach potential consumers
In the meta-model of knowledge sharing of search engines advertising has become a major revenue earning source for their sustenance
According to IAB standards sponsored search is the most dominant form of online advertising covering almost 43 of the entire market
Introduction Sponsored Search
The model of search engine advertising is more popularly known as Adwords
When a user queries a search engine a list of search results and sponsored results or advertisements are displayed
Advertisers bid on search keywords and pay the search engine according to Pay Per Click (PPC) model to display the advertisement on the query page containing the desired keywords
Introduction Problem StatementSearch keywords follow a long tail frequency distribution with
a small and fat head of highly frequent keywords
a long but thin tail of less frequent keywords
During the keyword auctions there is a high competition for head keywords while there is little to no competition for the tail keywords
This leads to underutilization of ad space of a large number of tail keywords Also it leads to ignorance of a diverse set of potential consumers who could be captured by targeting tail keywords
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Background Model of AdwordsThe present model is considered as
the online bipartite graph matching with advertisers as one disjoint set and incoming queries are the other
When a new query comes in it is to be matched to a set of advertisers
The advertisers are then ranked and their ads are display in that ranked order
Background Adwords Architecture
1 Analyze Query
2 Retrieve Relevant Advertisers
3 Bidding
4 Ranking Advertisers
Background Coverage Patterns- Central Idea
The basic idea of Coverage Pattern is inspired from the set cover problem in set theory
Given a universe U and a family S of subsets of U a cover is a subfamily C S subof sets whose union is U
Using the same notion coverage patterns aim to identify items that cover certain percentage of the entire data
A keypoint to be mentioned is that coverage patterns aim at identifying that usually ldquodo notrdquo occur together in contrast to frequent patterns that identify patterns in data that occur together
Background Coverage Patterns - Notations
Let W be a set of webpages of a website W = w1 w2 hellip wN
Let D be a set of transactions from the click stream data D = T1 T2 hellip such that T Wsub
X is defined as a patterns of webpages such that X W X = subwp wq hellip
Twi denotes the set of transactions containing the webpage wi and its cardinality is denoted as |Twi|
Background Coverage Patterns - Definitions
Background Coverage Patterns - Definitions
Background Coverage Patterns
A pattern is interesting if it has a high CS and low OR
A high CS value indicates more number of visitors and a low OR value means less repetitions amongst the visitors
A pattern is said to be interesting if CS(X) gt minCS(X) OR(X) lt maxOR and RF(wi) gt minRF
Background Coverage Patterns - Example
Dataset
Assuming minRF - 02 minCS - 03 and maxOR- to be 05
Ta is 5 Tb is 7 and f Tf is 1 So RF for a is 05 for b is 07 and for f is 01
RF(f) = 01 lt 02 (minRF) f will be removed RF(a) = 05 gt 02and RF(b) = 07 gt 02 so a and b are not removed
ba is a candidate pattern (Order in a pattern is decreasing order of the RF)
The Coverage Set for ba is 12345678910 and |CSet ba| is 10 Hence CS = 1010 = 1 gt 03 (minCS)
The transactions containing ba together are 110 and Ta = 5 so the overlap ratio is 25 = 04 lt maxOR is a coverage pattern
Since pattern ba has CS gt minCS and OR gt maxOR and a and b satisfy the minRF requirements ba is a coverage pattern
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Proposed Approach Basic IdeaBecause of the nature of distribution of search keywords there is very less
competition for tail keywords As a result there are very less or no advertisers for such keywords
We noticed that if could combine such keywords into groups such that these keyword groups have a certain number of visitors we can utilize the ad space of such keywords
To perform the grouping of search keywords we employ the notion of query taxonomy to group semantically similar words
These groups are further mined from the logs in the form of coverage patterns
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 4: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/4.jpg)
Introduction Sponsored Search
The model of search engine advertising is more popularly known as Adwords
When a user queries a search engine a list of search results and sponsored results or advertisements are displayed
Advertisers bid on search keywords and pay the search engine according to Pay Per Click (PPC) model to display the advertisement on the query page containing the desired keywords
Introduction Problem StatementSearch keywords follow a long tail frequency distribution with
a small and fat head of highly frequent keywords
a long but thin tail of less frequent keywords
During the keyword auctions there is a high competition for head keywords while there is little to no competition for the tail keywords
This leads to underutilization of ad space of a large number of tail keywords Also it leads to ignorance of a diverse set of potential consumers who could be captured by targeting tail keywords
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Background Model of AdwordsThe present model is considered as
the online bipartite graph matching with advertisers as one disjoint set and incoming queries are the other
When a new query comes in it is to be matched to a set of advertisers
The advertisers are then ranked and their ads are display in that ranked order
Background Adwords Architecture
1 Analyze Query
2 Retrieve Relevant Advertisers
3 Bidding
4 Ranking Advertisers
Background Coverage Patterns- Central Idea
The basic idea of Coverage Pattern is inspired from the set cover problem in set theory
Given a universe U and a family S of subsets of U a cover is a subfamily C S subof sets whose union is U
Using the same notion coverage patterns aim to identify items that cover certain percentage of the entire data
A keypoint to be mentioned is that coverage patterns aim at identifying that usually ldquodo notrdquo occur together in contrast to frequent patterns that identify patterns in data that occur together
Background Coverage Patterns - Notations
Let W be a set of webpages of a website W = w1 w2 hellip wN
Let D be a set of transactions from the click stream data D = T1 T2 hellip such that T Wsub
X is defined as a patterns of webpages such that X W X = subwp wq hellip
Twi denotes the set of transactions containing the webpage wi and its cardinality is denoted as |Twi|
Background Coverage Patterns - Definitions
Background Coverage Patterns - Definitions
Background Coverage Patterns
A pattern is interesting if it has a high CS and low OR
A high CS value indicates more number of visitors and a low OR value means less repetitions amongst the visitors
A pattern is said to be interesting if CS(X) gt minCS(X) OR(X) lt maxOR and RF(wi) gt minRF
Background Coverage Patterns - Example
Dataset
Assuming minRF - 02 minCS - 03 and maxOR- to be 05
Ta is 5 Tb is 7 and f Tf is 1 So RF for a is 05 for b is 07 and for f is 01
RF(f) = 01 lt 02 (minRF) f will be removed RF(a) = 05 gt 02and RF(b) = 07 gt 02 so a and b are not removed
ba is a candidate pattern (Order in a pattern is decreasing order of the RF)
The Coverage Set for ba is 12345678910 and |CSet ba| is 10 Hence CS = 1010 = 1 gt 03 (minCS)
The transactions containing ba together are 110 and Ta = 5 so the overlap ratio is 25 = 04 lt maxOR is a coverage pattern
Since pattern ba has CS gt minCS and OR gt maxOR and a and b satisfy the minRF requirements ba is a coverage pattern
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Proposed Approach Basic IdeaBecause of the nature of distribution of search keywords there is very less
competition for tail keywords As a result there are very less or no advertisers for such keywords
We noticed that if could combine such keywords into groups such that these keyword groups have a certain number of visitors we can utilize the ad space of such keywords
To perform the grouping of search keywords we employ the notion of query taxonomy to group semantically similar words
These groups are further mined from the logs in the form of coverage patterns
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 5: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/5.jpg)
Introduction Problem StatementSearch keywords follow a long tail frequency distribution with
a small and fat head of highly frequent keywords
a long but thin tail of less frequent keywords
During the keyword auctions there is a high competition for head keywords while there is little to no competition for the tail keywords
This leads to underutilization of ad space of a large number of tail keywords Also it leads to ignorance of a diverse set of potential consumers who could be captured by targeting tail keywords
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Background Model of AdwordsThe present model is considered as
the online bipartite graph matching with advertisers as one disjoint set and incoming queries are the other
When a new query comes in it is to be matched to a set of advertisers
The advertisers are then ranked and their ads are display in that ranked order
Background Adwords Architecture
1 Analyze Query
2 Retrieve Relevant Advertisers
3 Bidding
4 Ranking Advertisers
Background Coverage Patterns- Central Idea
The basic idea of Coverage Pattern is inspired from the set cover problem in set theory
Given a universe U and a family S of subsets of U a cover is a subfamily C S subof sets whose union is U
Using the same notion coverage patterns aim to identify items that cover certain percentage of the entire data
A keypoint to be mentioned is that coverage patterns aim at identifying that usually ldquodo notrdquo occur together in contrast to frequent patterns that identify patterns in data that occur together
Background Coverage Patterns - Notations
Let W be a set of webpages of a website W = w1 w2 hellip wN
Let D be a set of transactions from the click stream data D = T1 T2 hellip such that T Wsub
X is defined as a patterns of webpages such that X W X = subwp wq hellip
Twi denotes the set of transactions containing the webpage wi and its cardinality is denoted as |Twi|
Background Coverage Patterns - Definitions
Background Coverage Patterns - Definitions
Background Coverage Patterns
A pattern is interesting if it has a high CS and low OR
A high CS value indicates more number of visitors and a low OR value means less repetitions amongst the visitors
A pattern is said to be interesting if CS(X) gt minCS(X) OR(X) lt maxOR and RF(wi) gt minRF
Background Coverage Patterns - Example
Dataset
Assuming minRF - 02 minCS - 03 and maxOR- to be 05
Ta is 5 Tb is 7 and f Tf is 1 So RF for a is 05 for b is 07 and for f is 01
RF(f) = 01 lt 02 (minRF) f will be removed RF(a) = 05 gt 02and RF(b) = 07 gt 02 so a and b are not removed
ba is a candidate pattern (Order in a pattern is decreasing order of the RF)
The Coverage Set for ba is 12345678910 and |CSet ba| is 10 Hence CS = 1010 = 1 gt 03 (minCS)
The transactions containing ba together are 110 and Ta = 5 so the overlap ratio is 25 = 04 lt maxOR is a coverage pattern
Since pattern ba has CS gt minCS and OR gt maxOR and a and b satisfy the minRF requirements ba is a coverage pattern
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Proposed Approach Basic IdeaBecause of the nature of distribution of search keywords there is very less
competition for tail keywords As a result there are very less or no advertisers for such keywords
We noticed that if could combine such keywords into groups such that these keyword groups have a certain number of visitors we can utilize the ad space of such keywords
To perform the grouping of search keywords we employ the notion of query taxonomy to group semantically similar words
These groups are further mined from the logs in the form of coverage patterns
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 6: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/6.jpg)
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Background Model of AdwordsThe present model is considered as
the online bipartite graph matching with advertisers as one disjoint set and incoming queries are the other
When a new query comes in it is to be matched to a set of advertisers
The advertisers are then ranked and their ads are display in that ranked order
Background Adwords Architecture
1 Analyze Query
2 Retrieve Relevant Advertisers
3 Bidding
4 Ranking Advertisers
Background Coverage Patterns- Central Idea
The basic idea of Coverage Pattern is inspired from the set cover problem in set theory
Given a universe U and a family S of subsets of U a cover is a subfamily C S subof sets whose union is U
Using the same notion coverage patterns aim to identify items that cover certain percentage of the entire data
A keypoint to be mentioned is that coverage patterns aim at identifying that usually ldquodo notrdquo occur together in contrast to frequent patterns that identify patterns in data that occur together
Background Coverage Patterns - Notations
Let W be a set of webpages of a website W = w1 w2 hellip wN
Let D be a set of transactions from the click stream data D = T1 T2 hellip such that T Wsub
X is defined as a patterns of webpages such that X W X = subwp wq hellip
Twi denotes the set of transactions containing the webpage wi and its cardinality is denoted as |Twi|
Background Coverage Patterns - Definitions
Background Coverage Patterns - Definitions
Background Coverage Patterns
A pattern is interesting if it has a high CS and low OR
A high CS value indicates more number of visitors and a low OR value means less repetitions amongst the visitors
A pattern is said to be interesting if CS(X) gt minCS(X) OR(X) lt maxOR and RF(wi) gt minRF
Background Coverage Patterns - Example
Dataset
Assuming minRF - 02 minCS - 03 and maxOR- to be 05
Ta is 5 Tb is 7 and f Tf is 1 So RF for a is 05 for b is 07 and for f is 01
RF(f) = 01 lt 02 (minRF) f will be removed RF(a) = 05 gt 02and RF(b) = 07 gt 02 so a and b are not removed
ba is a candidate pattern (Order in a pattern is decreasing order of the RF)
The Coverage Set for ba is 12345678910 and |CSet ba| is 10 Hence CS = 1010 = 1 gt 03 (minCS)
The transactions containing ba together are 110 and Ta = 5 so the overlap ratio is 25 = 04 lt maxOR is a coverage pattern
Since pattern ba has CS gt minCS and OR gt maxOR and a and b satisfy the minRF requirements ba is a coverage pattern
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Proposed Approach Basic IdeaBecause of the nature of distribution of search keywords there is very less
competition for tail keywords As a result there are very less or no advertisers for such keywords
We noticed that if could combine such keywords into groups such that these keyword groups have a certain number of visitors we can utilize the ad space of such keywords
To perform the grouping of search keywords we employ the notion of query taxonomy to group semantically similar words
These groups are further mined from the logs in the form of coverage patterns
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 7: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/7.jpg)
Background Model of AdwordsThe present model is considered as
the online bipartite graph matching with advertisers as one disjoint set and incoming queries are the other
When a new query comes in it is to be matched to a set of advertisers
The advertisers are then ranked and their ads are display in that ranked order
Background Adwords Architecture
1 Analyze Query
2 Retrieve Relevant Advertisers
3 Bidding
4 Ranking Advertisers
Background Coverage Patterns- Central Idea
The basic idea of Coverage Pattern is inspired from the set cover problem in set theory
Given a universe U and a family S of subsets of U a cover is a subfamily C S subof sets whose union is U
Using the same notion coverage patterns aim to identify items that cover certain percentage of the entire data
A keypoint to be mentioned is that coverage patterns aim at identifying that usually ldquodo notrdquo occur together in contrast to frequent patterns that identify patterns in data that occur together
Background Coverage Patterns - Notations
Let W be a set of webpages of a website W = w1 w2 hellip wN
Let D be a set of transactions from the click stream data D = T1 T2 hellip such that T Wsub
X is defined as a patterns of webpages such that X W X = subwp wq hellip
Twi denotes the set of transactions containing the webpage wi and its cardinality is denoted as |Twi|
Background Coverage Patterns - Definitions
Background Coverage Patterns - Definitions
Background Coverage Patterns
A pattern is interesting if it has a high CS and low OR
A high CS value indicates more number of visitors and a low OR value means less repetitions amongst the visitors
A pattern is said to be interesting if CS(X) gt minCS(X) OR(X) lt maxOR and RF(wi) gt minRF
Background Coverage Patterns - Example
Dataset
Assuming minRF - 02 minCS - 03 and maxOR- to be 05
Ta is 5 Tb is 7 and f Tf is 1 So RF for a is 05 for b is 07 and for f is 01
RF(f) = 01 lt 02 (minRF) f will be removed RF(a) = 05 gt 02and RF(b) = 07 gt 02 so a and b are not removed
ba is a candidate pattern (Order in a pattern is decreasing order of the RF)
The Coverage Set for ba is 12345678910 and |CSet ba| is 10 Hence CS = 1010 = 1 gt 03 (minCS)
The transactions containing ba together are 110 and Ta = 5 so the overlap ratio is 25 = 04 lt maxOR is a coverage pattern
Since pattern ba has CS gt minCS and OR gt maxOR and a and b satisfy the minRF requirements ba is a coverage pattern
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Proposed Approach Basic IdeaBecause of the nature of distribution of search keywords there is very less
competition for tail keywords As a result there are very less or no advertisers for such keywords
We noticed that if could combine such keywords into groups such that these keyword groups have a certain number of visitors we can utilize the ad space of such keywords
To perform the grouping of search keywords we employ the notion of query taxonomy to group semantically similar words
These groups are further mined from the logs in the form of coverage patterns
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 8: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/8.jpg)
Background Adwords Architecture
1 Analyze Query
2 Retrieve Relevant Advertisers
3 Bidding
4 Ranking Advertisers
Background Coverage Patterns- Central Idea
The basic idea of Coverage Pattern is inspired from the set cover problem in set theory
Given a universe U and a family S of subsets of U a cover is a subfamily C S subof sets whose union is U
Using the same notion coverage patterns aim to identify items that cover certain percentage of the entire data
A keypoint to be mentioned is that coverage patterns aim at identifying that usually ldquodo notrdquo occur together in contrast to frequent patterns that identify patterns in data that occur together
Background Coverage Patterns - Notations
Let W be a set of webpages of a website W = w1 w2 hellip wN
Let D be a set of transactions from the click stream data D = T1 T2 hellip such that T Wsub
X is defined as a patterns of webpages such that X W X = subwp wq hellip
Twi denotes the set of transactions containing the webpage wi and its cardinality is denoted as |Twi|
Background Coverage Patterns - Definitions
Background Coverage Patterns - Definitions
Background Coverage Patterns
A pattern is interesting if it has a high CS and low OR
A high CS value indicates more number of visitors and a low OR value means less repetitions amongst the visitors
A pattern is said to be interesting if CS(X) gt minCS(X) OR(X) lt maxOR and RF(wi) gt minRF
Background Coverage Patterns - Example
Dataset
Assuming minRF - 02 minCS - 03 and maxOR- to be 05
Ta is 5 Tb is 7 and f Tf is 1 So RF for a is 05 for b is 07 and for f is 01
RF(f) = 01 lt 02 (minRF) f will be removed RF(a) = 05 gt 02and RF(b) = 07 gt 02 so a and b are not removed
ba is a candidate pattern (Order in a pattern is decreasing order of the RF)
The Coverage Set for ba is 12345678910 and |CSet ba| is 10 Hence CS = 1010 = 1 gt 03 (minCS)
The transactions containing ba together are 110 and Ta = 5 so the overlap ratio is 25 = 04 lt maxOR is a coverage pattern
Since pattern ba has CS gt minCS and OR gt maxOR and a and b satisfy the minRF requirements ba is a coverage pattern
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Proposed Approach Basic IdeaBecause of the nature of distribution of search keywords there is very less
competition for tail keywords As a result there are very less or no advertisers for such keywords
We noticed that if could combine such keywords into groups such that these keyword groups have a certain number of visitors we can utilize the ad space of such keywords
To perform the grouping of search keywords we employ the notion of query taxonomy to group semantically similar words
These groups are further mined from the logs in the form of coverage patterns
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 9: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/9.jpg)
Background Coverage Patterns- Central Idea
The basic idea of Coverage Pattern is inspired from the set cover problem in set theory
Given a universe U and a family S of subsets of U a cover is a subfamily C S subof sets whose union is U
Using the same notion coverage patterns aim to identify items that cover certain percentage of the entire data
A keypoint to be mentioned is that coverage patterns aim at identifying that usually ldquodo notrdquo occur together in contrast to frequent patterns that identify patterns in data that occur together
Background Coverage Patterns - Notations
Let W be a set of webpages of a website W = w1 w2 hellip wN
Let D be a set of transactions from the click stream data D = T1 T2 hellip such that T Wsub
X is defined as a patterns of webpages such that X W X = subwp wq hellip
Twi denotes the set of transactions containing the webpage wi and its cardinality is denoted as |Twi|
Background Coverage Patterns - Definitions
Background Coverage Patterns - Definitions
Background Coverage Patterns
A pattern is interesting if it has a high CS and low OR
A high CS value indicates more number of visitors and a low OR value means less repetitions amongst the visitors
A pattern is said to be interesting if CS(X) gt minCS(X) OR(X) lt maxOR and RF(wi) gt minRF
Background Coverage Patterns - Example
Dataset
Assuming minRF - 02 minCS - 03 and maxOR- to be 05
Ta is 5 Tb is 7 and f Tf is 1 So RF for a is 05 for b is 07 and for f is 01
RF(f) = 01 lt 02 (minRF) f will be removed RF(a) = 05 gt 02and RF(b) = 07 gt 02 so a and b are not removed
ba is a candidate pattern (Order in a pattern is decreasing order of the RF)
The Coverage Set for ba is 12345678910 and |CSet ba| is 10 Hence CS = 1010 = 1 gt 03 (minCS)
The transactions containing ba together are 110 and Ta = 5 so the overlap ratio is 25 = 04 lt maxOR is a coverage pattern
Since pattern ba has CS gt minCS and OR gt maxOR and a and b satisfy the minRF requirements ba is a coverage pattern
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Proposed Approach Basic IdeaBecause of the nature of distribution of search keywords there is very less
competition for tail keywords As a result there are very less or no advertisers for such keywords
We noticed that if could combine such keywords into groups such that these keyword groups have a certain number of visitors we can utilize the ad space of such keywords
To perform the grouping of search keywords we employ the notion of query taxonomy to group semantically similar words
These groups are further mined from the logs in the form of coverage patterns
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 10: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/10.jpg)
Background Coverage Patterns - Notations
Let W be a set of webpages of a website W = w1 w2 hellip wN
Let D be a set of transactions from the click stream data D = T1 T2 hellip such that T Wsub
X is defined as a patterns of webpages such that X W X = subwp wq hellip
Twi denotes the set of transactions containing the webpage wi and its cardinality is denoted as |Twi|
Background Coverage Patterns - Definitions
Background Coverage Patterns - Definitions
Background Coverage Patterns
A pattern is interesting if it has a high CS and low OR
A high CS value indicates more number of visitors and a low OR value means less repetitions amongst the visitors
A pattern is said to be interesting if CS(X) gt minCS(X) OR(X) lt maxOR and RF(wi) gt minRF
Background Coverage Patterns - Example
Dataset
Assuming minRF - 02 minCS - 03 and maxOR- to be 05
Ta is 5 Tb is 7 and f Tf is 1 So RF for a is 05 for b is 07 and for f is 01
RF(f) = 01 lt 02 (minRF) f will be removed RF(a) = 05 gt 02and RF(b) = 07 gt 02 so a and b are not removed
ba is a candidate pattern (Order in a pattern is decreasing order of the RF)
The Coverage Set for ba is 12345678910 and |CSet ba| is 10 Hence CS = 1010 = 1 gt 03 (minCS)
The transactions containing ba together are 110 and Ta = 5 so the overlap ratio is 25 = 04 lt maxOR is a coverage pattern
Since pattern ba has CS gt minCS and OR gt maxOR and a and b satisfy the minRF requirements ba is a coverage pattern
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Proposed Approach Basic IdeaBecause of the nature of distribution of search keywords there is very less
competition for tail keywords As a result there are very less or no advertisers for such keywords
We noticed that if could combine such keywords into groups such that these keyword groups have a certain number of visitors we can utilize the ad space of such keywords
To perform the grouping of search keywords we employ the notion of query taxonomy to group semantically similar words
These groups are further mined from the logs in the form of coverage patterns
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 11: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/11.jpg)
Background Coverage Patterns - Definitions
Background Coverage Patterns - Definitions
Background Coverage Patterns
A pattern is interesting if it has a high CS and low OR
A high CS value indicates more number of visitors and a low OR value means less repetitions amongst the visitors
A pattern is said to be interesting if CS(X) gt minCS(X) OR(X) lt maxOR and RF(wi) gt minRF
Background Coverage Patterns - Example
Dataset
Assuming minRF - 02 minCS - 03 and maxOR- to be 05
Ta is 5 Tb is 7 and f Tf is 1 So RF for a is 05 for b is 07 and for f is 01
RF(f) = 01 lt 02 (minRF) f will be removed RF(a) = 05 gt 02and RF(b) = 07 gt 02 so a and b are not removed
ba is a candidate pattern (Order in a pattern is decreasing order of the RF)
The Coverage Set for ba is 12345678910 and |CSet ba| is 10 Hence CS = 1010 = 1 gt 03 (minCS)
The transactions containing ba together are 110 and Ta = 5 so the overlap ratio is 25 = 04 lt maxOR is a coverage pattern
Since pattern ba has CS gt minCS and OR gt maxOR and a and b satisfy the minRF requirements ba is a coverage pattern
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Proposed Approach Basic IdeaBecause of the nature of distribution of search keywords there is very less
competition for tail keywords As a result there are very less or no advertisers for such keywords
We noticed that if could combine such keywords into groups such that these keyword groups have a certain number of visitors we can utilize the ad space of such keywords
To perform the grouping of search keywords we employ the notion of query taxonomy to group semantically similar words
These groups are further mined from the logs in the form of coverage patterns
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 12: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/12.jpg)
Background Coverage Patterns - Definitions
Background Coverage Patterns
A pattern is interesting if it has a high CS and low OR
A high CS value indicates more number of visitors and a low OR value means less repetitions amongst the visitors
A pattern is said to be interesting if CS(X) gt minCS(X) OR(X) lt maxOR and RF(wi) gt minRF
Background Coverage Patterns - Example
Dataset
Assuming minRF - 02 minCS - 03 and maxOR- to be 05
Ta is 5 Tb is 7 and f Tf is 1 So RF for a is 05 for b is 07 and for f is 01
RF(f) = 01 lt 02 (minRF) f will be removed RF(a) = 05 gt 02and RF(b) = 07 gt 02 so a and b are not removed
ba is a candidate pattern (Order in a pattern is decreasing order of the RF)
The Coverage Set for ba is 12345678910 and |CSet ba| is 10 Hence CS = 1010 = 1 gt 03 (minCS)
The transactions containing ba together are 110 and Ta = 5 so the overlap ratio is 25 = 04 lt maxOR is a coverage pattern
Since pattern ba has CS gt minCS and OR gt maxOR and a and b satisfy the minRF requirements ba is a coverage pattern
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Proposed Approach Basic IdeaBecause of the nature of distribution of search keywords there is very less
competition for tail keywords As a result there are very less or no advertisers for such keywords
We noticed that if could combine such keywords into groups such that these keyword groups have a certain number of visitors we can utilize the ad space of such keywords
To perform the grouping of search keywords we employ the notion of query taxonomy to group semantically similar words
These groups are further mined from the logs in the form of coverage patterns
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 13: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/13.jpg)
Background Coverage Patterns
A pattern is interesting if it has a high CS and low OR
A high CS value indicates more number of visitors and a low OR value means less repetitions amongst the visitors
A pattern is said to be interesting if CS(X) gt minCS(X) OR(X) lt maxOR and RF(wi) gt minRF
Background Coverage Patterns - Example
Dataset
Assuming minRF - 02 minCS - 03 and maxOR- to be 05
Ta is 5 Tb is 7 and f Tf is 1 So RF for a is 05 for b is 07 and for f is 01
RF(f) = 01 lt 02 (minRF) f will be removed RF(a) = 05 gt 02and RF(b) = 07 gt 02 so a and b are not removed
ba is a candidate pattern (Order in a pattern is decreasing order of the RF)
The Coverage Set for ba is 12345678910 and |CSet ba| is 10 Hence CS = 1010 = 1 gt 03 (minCS)
The transactions containing ba together are 110 and Ta = 5 so the overlap ratio is 25 = 04 lt maxOR is a coverage pattern
Since pattern ba has CS gt minCS and OR gt maxOR and a and b satisfy the minRF requirements ba is a coverage pattern
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Proposed Approach Basic IdeaBecause of the nature of distribution of search keywords there is very less
competition for tail keywords As a result there are very less or no advertisers for such keywords
We noticed that if could combine such keywords into groups such that these keyword groups have a certain number of visitors we can utilize the ad space of such keywords
To perform the grouping of search keywords we employ the notion of query taxonomy to group semantically similar words
These groups are further mined from the logs in the form of coverage patterns
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 14: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/14.jpg)
Background Coverage Patterns - Example
Dataset
Assuming minRF - 02 minCS - 03 and maxOR- to be 05
Ta is 5 Tb is 7 and f Tf is 1 So RF for a is 05 for b is 07 and for f is 01
RF(f) = 01 lt 02 (minRF) f will be removed RF(a) = 05 gt 02and RF(b) = 07 gt 02 so a and b are not removed
ba is a candidate pattern (Order in a pattern is decreasing order of the RF)
The Coverage Set for ba is 12345678910 and |CSet ba| is 10 Hence CS = 1010 = 1 gt 03 (minCS)
The transactions containing ba together are 110 and Ta = 5 so the overlap ratio is 25 = 04 lt maxOR is a coverage pattern
Since pattern ba has CS gt minCS and OR gt maxOR and a and b satisfy the minRF requirements ba is a coverage pattern
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Proposed Approach Basic IdeaBecause of the nature of distribution of search keywords there is very less
competition for tail keywords As a result there are very less or no advertisers for such keywords
We noticed that if could combine such keywords into groups such that these keyword groups have a certain number of visitors we can utilize the ad space of such keywords
To perform the grouping of search keywords we employ the notion of query taxonomy to group semantically similar words
These groups are further mined from the logs in the form of coverage patterns
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 15: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/15.jpg)
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Proposed Approach Basic IdeaBecause of the nature of distribution of search keywords there is very less
competition for tail keywords As a result there are very less or no advertisers for such keywords
We noticed that if could combine such keywords into groups such that these keyword groups have a certain number of visitors we can utilize the ad space of such keywords
To perform the grouping of search keywords we employ the notion of query taxonomy to group semantically similar words
These groups are further mined from the logs in the form of coverage patterns
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 16: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/16.jpg)
Proposed Approach Basic IdeaBecause of the nature of distribution of search keywords there is very less
competition for tail keywords As a result there are very less or no advertisers for such keywords
We noticed that if could combine such keywords into groups such that these keyword groups have a certain number of visitors we can utilize the ad space of such keywords
To perform the grouping of search keywords we employ the notion of query taxonomy to group semantically similar words
These groups are further mined from the logs in the form of coverage patterns
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 17: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/17.jpg)
Proposed ModelWe propose to add a middle layer of
coverage patterns to the bipartite model of Adwords
In the proposed model the incoming queries are first matched to a coverage pattern using the concept taxonomy
The coverage pattern is then matched to a set of advertisers
The advertisers are then ranked according to their bids and the ads are displayed
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 18: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/18.jpg)
Proposed Architecture
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 19: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/19.jpg)
Architecture ComparisonStep Modification with respect to bipartite
architecture
Analyze Query This step remains the same except that the subconcept of the query is also retrieved
Retrieve RelevantAds From the
Matching
In this step the advertisers who have been matched to the coverage pattern containing the
subconcept of the query are retrieved (The matching between coverage patterns and
advertisers would be explained later)
Bidding Stays the same
Ranking Advertisers Stays the same
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 20: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/20.jpg)
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 21: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/21.jpg)
Coverage Pattern and Advertisers MatchingCoverage Pattern and Advertiser matching is the most important phase in the
architecture
This has been further divided into four steps
a Converting Query Logs to Concept Transactions
b Extraction of Coverage Patterns
c Estimation of Number of Impressions for Advertisers
d Matching Coverage Patterns and Advertisers
Each step is explained in the later slides
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 22: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/22.jpg)
Step 1 Converting Query Logs to Concept Transactions
One key point to note here is that the web query logs cannot be directly mined for coverage patterns because of the large vocabulary size (even when we only consider English)
To generalize the coverage pattern mining we proposed to use a three-level concept taxonomy to classify queries into a pair of concept and subconcept
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 23: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/23.jpg)
Step 1Converting Query Logs to Concept Transactions (cntd)
Using the same taxonomy we convert the web query logs into concept transactions using the techniques of query classification
To define a transaction we consider a session boundary of 30 mins in the query logs for each user
Sample Sessions
Converted Concepted Transactions
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 24: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/24.jpg)
Step 2 Extraction of Coverage Patterns
Coverage patterns are extracted from the converted concept transactions
One key point to note is that coverage patterns mine unique visitors while the standard models of advertising are either based on Impressions or Clicks
So we convert coverage patternsrsquo coverage into number of impressions as follows
For the above example we consider the concept of Science and Agriculture Biology Chemistry Environment Physics and Technology as its subconcepts
The transaction size is also assumed to be 1000 NOTE We also rank the coverage patterns in
ascending order of their CS - OR parameter
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 25: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/25.jpg)
Step 3 Estimating Required Impressions for Advertisers
In Adwords advertisers create an ad campaign for their website
In an ad campaign a daily budget and a bid is specified on the keywords that they chose to bid upon
Using CTR bid and daily budget values we calculated the number of impressions it will take to exhaust the budget of an advertiser using the following identity
The above table shows details of nine advertisers who bid upon the concept of Science
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 26: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/26.jpg)
Step 4 Matching Advertisers to Coverage Patterns
With coverage patterns and advertisers in the same unit of number of impressions we can create a matching between the two
The matching can be termed as a MANY-TO-ONE matching between coverage patterns and advertisers because a coverage patterns covers multiple keywords from different nodes in the taxonomy
The matching algorithm has a relaxation parameter ε to perform faster
The algorithms loops over coverage patterns and then advertisers and a coverage pattern is allocated to an advertisers if the following condition is satisfied
AdImpressions - 1048576 CPCoverage lt ε times AdImpressions
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 27: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/27.jpg)
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 28: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/28.jpg)
Experiments DatasetWe performed a comparative study on the bipartite model of Adwords with and
without coverage patterns layers
We used AOL search query dataset to run the experiments
We took the most popular four categories of queries to run our experiments
Query Dataset for the most popular four categories
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 29: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/29.jpg)
Experiments Performance Metrics
1 Number of Advertisements per Session (AS) as the ratio of Sum of Unique Advertisements of all Sessions (SUAS) and Number of Sessions with Advertisements (NSA) to indicate the utilization of a session Higher value of AS indicates better use of ad space
AS = SUAS NAS
2 An increase in diversity among the viewers of the advertisements was also observed To indicate the same the value of Sessions per Advertisement (SA) which is the ratio of Number of Advertisements of all Sessions (NAS) to Number of Advertisements (NA) Higher value of metric implies the more number of unique eyeballs and thus increasing the chances of the advertisement being clicked by diverse users
SA = NASNA
Experiments Performance Metrics
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 30: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/30.jpg)
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results- Utilization of Ad Space
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 31: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/31.jpg)
Graphs show a comparison of the bipartite Adwords system - With and Without Coverage Patterns layer
Experiments Results - Diversity
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 32: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/32.jpg)
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 33: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/33.jpg)
Related WorkMost works in Adwords target algorithms to optimize different aspects of the
system including revenue welfarism and display of ads
Another aspect that is targeted in Adwords is the bidding scenarios with respect to Adwords Several studies have touched up on how to increase revenue in dynamic bidding scenario when you only have a partial information of the system
In this paper we have proposed an architectural solution to use the ad space of the search keywords Bidding strategies and Budget optimization can be placed on top of it
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 34: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/34.jpg)
Outline
Introduction
Background
Proposed Approach
Experiments
Related Work
Conclusions and Future Work
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-
![Page 35: An Approach to cover more advertisers in Adwords](https://reader036.vdocument.in/reader036/viewer/2022070513/588537551a28ab26518b618f/html5/thumbnails/35.jpg)
Conclusions and Future Work
In this paper an architectural solution is proposed for Adwords to utilize the ad space of tail keywords The proposed approach also considerable improvement with respect to diversity in reach of advertisements
We plan on investigating the coverage patterns approach with respect to different taxonomies We believe a hybrid taxonomy would be the best when it comes to Adwords architecture
We also plan on expanding the user boundaries of exploration in searching beyond search sessions We plan on extracting user goals and modelling the transactions from them
- An Approach to Cover More Advertisers in Adwords
- Outline
- Introduction
- Introduction Sponsored Search
- Introduction Problem Statement
- Outline (2)
- Background Model of Adwords
- Background Adwords Architecture
- Background Coverage Patterns- Central Idea
- Background Coverage Patterns - Notations
- Background Coverage Patterns - Definitions
- Background Coverage Patterns - Definitions
- Background Coverage Patterns
- Background Coverage Patterns - Example
- Outline (3)
- Proposed Approach Basic Idea
- Proposed Model
- Proposed Architecture
- Architecture Comparison
- Outline (4)
- Coverage Pattern and Advertisers Matching
- Step 1 Converting Query Logs to Concept Transactions
- Step 1Converting Query Logs to Concept Transactions (cntd)
- Step 2 Extraction of Coverage Patterns
- Step 3 Estimating Required Impressions for Advertisers
- Step 4 Matching Advertisers to Coverage Patterns
- Outline (5)
- Experiments Dataset
- Experiments Performance Metrics (2)
- Experiments Results- Utilization of Ad Space
- Experiments Results - Diversity
- Outline (6)
- Related Work
- Outline (7)
- Conclusions and Future Work
-