knowledge discovery and data mining evgueni smirnov
TRANSCRIPT
![Page 1: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/1.jpg)
Knowledge Discovery andKnowledge Discovery andData MiningData Mining
Evgueni Smirnov
![Page 2: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/2.jpg)
Outline
• Data Flood• Definition of Knowledge Discovery and
Data Mining• Possible Tasks:
– Classification Task– Regression Task– Clustering Task– Association-Rule Task
![Page 4: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/4.jpg)
Trends Leading to Data Flood
• Moore’s law– Computer Speed doubles every 18 months
• Storage law– total storage doubles every 9 months
As a result:• More data is captured:
– Storage technology faster and cheaper
– DBMS capable of handling bigger DB
![Page 5: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/5.jpg)
Trends Leading to Data Flood
• More data is generated:– Business:
• Supermarket chains• Banks, • Telecoms,• E-commerce, etc.
– Web– Science:
• astronomy, • physics, • biology,• medicine etc.
![Page 6: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/6.jpg)
Consequence
• Very little data will ever be looked at by a human, and thus, we need to automate the process of Knowledge Discovery to make sense and use of data.
![Page 7: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/7.jpg)
Definition of Knowledge Discovery
• Knowledge Discovery in Data is non-trivial process of identifying – valid– novel– potentially useful– and ultimately understandable patterns in data.
• from Advances in Knowledge Discovery and Data Mining, Fayyad, Piatetsky-Shapiro, Smyth, and Uthurusamy, (Chapter 1), AAAI/MIT Press 1996.
![Page 8: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/8.jpg)
Related Fields
Statistics
MachineLearning
Databases
Visualization
Knowledge Discovery
![Page 9: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/9.jpg)
Knowledge-Discovery Methodology
data
Targetdata
Processeddata
Transformeddata Patterns
Knowledge
Selection
Preprocessing& cleaning
Transformation& featureselection
Data Mining
InterpretationEvaluation
Data Mining is searching for patterns of interest in a particular representation.
![Page 10: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/10.jpg)
Data-Mining Tasks
• Classification Task
• Regression Task
• Clustering Task
• Association-Rule Task
![Page 11: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/11.jpg)
Classification Task
• Given: a collection of instances (training set)– Each instances is represented by a set of attributes, one of
the attributes is the class attribute.
• Find: a classifier for the class attribute as a function of the values of other attributes.
• Goal: previously unseen instances should be assigned a class as accurately as possible.
![Page 12: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/12.jpg)
Example 1
Tid Refund MaritalStatus
TaxableIncome Cheat
1 Yes Single 125K No
2 No Married 100K No
3 No Single 70K No
4 Yes Married 120K No
5 No Divorced 95K Yes
6 No Married 60K No
7 Yes Divorced 220K No
8 No Single 85K Yes
9 No Married 75K No
10 No Single 90K Yes10
categoric
al
categoric
al
continuous
class
Refund MaritalStatus
TaxableIncome Cheat
No Single 75K ?
Yes Married 50K ?
No Married 150K ?
Yes Divorced 90K ?
No Single 40K ?
No Married 80K ?10
TestSet
Training Set
ClassifierLearn
Classifier
![Page 13: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/13.jpg)
Example 2
• Fraud Detection– Goal: Predict fraudulent cases in credit card
transactions.– Approach:
• Use credit card transactions and the information on its account-holder as attributes.
– When does a customer buy, what does he buy, how often he pays on time, etc
• Label past transactions as fraud or fair transactions. This forms the class attribute.
• Learn a model for the class of the transactions.• Use this model to detect fraud by observing credit card
transactions on an account.
![Page 14: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/14.jpg)
Regression Task
• Predict a value of a given continuous valued variable based on the values of other variables, assuming a linear or nonlinear model of dependency.
Examples:• Predicting sales amounts of new product based on advertising expenditure.• Predicting wind velocities as a function of temperature, humidity, air pressure, etc.• Time series prediction of stock market indices.
![Page 15: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/15.jpg)
Clustering Task
• Given a set of data points, each having a set of attributes, and a similarity measure among them, find clusters such that:– Data points in one cluster are more similar;
– Data points in separate clusters are less similar.
Intra-cluster distancesare minimized
Intra-cluster distancesare minimized
Inter-cluster distancesare maximized
Inter-cluster distancesare maximized
![Page 16: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/16.jpg)
Example
• Market Segmentation:– Goal: subdivide a market into distinct subsets of
customers where any subset may conceivably be selected as a market target to be reached with a distinct marketing mix.
– Approach: • Collect different attributes of customers based on their
geographical and lifestyle related information.• Find clusters of similar customers.• Measure the clustering quality by observing buying patterns
of customers in same cluster vs. those from different clusters.
![Page 17: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/17.jpg)
Association-Rule Task
• Given a set of records each of which contain some number of items from a given collection;– Produce dependency rules which will predict
occurrence of an item based on occurrences of other items.
TID Items
1 Bread, Coke, Milk
2 Beer, Bread
3 Beer, Coke, Diaper, Milk
4 Beer, Bread, Diaper, Milk
5 Coke, Diaper, Milk
Rules Discovered: Milk --> Coke Diaper, Milk --> Beer
Rules Discovered: Milk --> Coke Diaper, Milk --> Beer
![Page 18: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/18.jpg)
Example
• Supermarket shelf management.– Goal: To identify items that are bought together by
sufficiently many customers.
– Approach: Process the point-of-sale data collected with barcode scanners to find dependencies among items.
– A classic rule --• If a customer buys diaper and milk, then he is very likely to
buy beer.
• So, don’t be surprised if you find six-packs stacked next to diapers!
![Page 19: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/19.jpg)
Course Overview
data
Processeddata
Selection
Preprocessing& cleaning
Transformation& featureselection
Data Mining
InterpretationEvaluation
Monday: Decision Trees and Decision Rules (E. Smirnov)
Evaluation of learning models (E. Smirnov)
![Page 20: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/20.jpg)
Course Overview
data
Processeddata
Selection
Preprocessing& cleaning
Transformation& featureselection
Data Mining
InterpretationEvaluation
Tuesday: Feature selection and extraction (G. Nalbantov)
Regression (G. Nalbantov)
![Page 21: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/21.jpg)
Course Overview
data
Processeddata
Selection
Preprocessing& cleaning
Transformation& featureselection
Data Mining
InterpretationEvaluation
Wednesday : Instance learning and Bayesian learning (E. Smirnov)
Association Rules (E. Smirnov)
![Page 22: Knowledge Discovery and Data Mining Evgueni Smirnov](https://reader036.vdocument.in/reader036/viewer/2022062718/56649e975503460f94b9ab61/html5/thumbnails/22.jpg)
Course Overview
data
Processeddata
Selection
Preprocessing& cleaning
Transformation& featureselection
Data Mining
InterpretationEvaluation
Thursday: Support Vector Machines (G. Nalbantov)