Deriving Concept-Based User Profilesfrom Search Engine Logs

To order this project call 020-30858066/9970119370

Apple store

iPodiPhone

Apple

Apple Grower

Apple Farm Market

Apple

Problem Statement

User 1

User 2

User profiling is a fundamental component of any personalization applications. Most existing user profiling strategies are based on objects that users are interested in (i.e., positive preferences), but not the objects that users dislike (i.e., negative preferences).

Concept-Based Clustering: A major problem of current Web search is that search queries are usually short and ambiguous, and thus are insufficient for specifying the precise user needs.

Concept-Based Clustering: To alleviate this problem, we introduce an effective approach that captures the user’s conceptual preferences in order to provide personalized search results.

1. First, we develop online techniques that extract concepts from the web-snippets of the search result returned from a query and use the concepts to identify related queries for that query.

2. Second, we propose a new two phase personalized clustering algorithm that is able to generate personalized query clusters.

Also for personalization of user profiles, we focus on search engine personalization and develop several concept-based user profiling methods that are based on both positive and negative preferences.

The proposed framework contains two components

Concept-Based Clustering:

1. To evaluate the effectiveness of our technique, a Google middleware was developed for collecting click-through data to conduct experimental evaluation.

2. Experimental results show that our approach has better precision and recall than the existing query clustering methods.

3. We evaluate the proposed methods against our query clustering method. Experimental results show that profiles which capture and utilize both of the user’s positive and negative preferences perform the best.

User Profiling:

1. An important result from the experiments is that profiles with negative preferences can increase the separation between similar and dissimilar queries.

2. The separation provides a clear threshold for an clustering algorithm to terminate and improve the overall quality of the resulting query clusters.

How we will evaluate resultant systems performance?

Existing System

MOST commercial search engines return roughly the same results for the same query, regardless of the user’s real interest. Since queries submitted to search engines tend to be short and ambiguous, they are not likely to be able to express the user’s precise needs.

For example, a farmer may use the query “apple” to find information about growing delicious apples, while graphic designers may use the same query to find information about Apple Computer.

As a result, lots of pages retrieved may be irrelevant to the user needs because of the ambiguous queries. On the other hand, users may not want to reformulate their queries using more search terms, since it imposes additional burden on them during searching.

To increase the relevance of search results, personalized search engines create user profiles to capture the users’ personal preferences and as such identify the actual goal of the input query.

Proposed System has focused on the automatic learning of user preferences from users’ search histories or browsed documents and the development of personalized systems based on the learned user preferences.

The underlying idea of our proposed technique is based on concepts and their relations extracted from the submitted user queries, the web-snippets and the click-through data. Click-through data was exploited in the personalized clustering process to identify user preferences.

Proposed System

Proposed Work-Flow

Fig. User Profiling from Concept Based Clustering

Algorithm Used:

Fig. Performing personalized concept-based clustering algorithm on a small set of click-through data. Starting from top left: (a) The original bipartite graph. (b), (c) Initial clustering. (d), (e) Community merging

Processor - Pentium –III Speed - 1.1 Ghz RAM - 256 MB(min) Hard Disk - 20 GB Floppy Drive - 1.44 MB Key Board - Standard Windows

Keyboard Mouse - Two or Three Button Mouse Monitor - SVGA

H/W System Configuration

Operating System - Windows95/98/2000/XP Application Server - Tomcat5.0/6.X Front End - HTML, JAVA Script - JavaScript Server side Script - JSP, Servlets Database - Microsoft Access

Database Database Connectivity - JDBC

S/W System Configuration

An accurate user profile can greatly improve a search engine’s performance by identifying the information needs for individual users.

The techniques make use of click-through data to extract from Web-snippets to build concept-based user profiles automatically.

Apart from improving the quality of the resulting clusters, the negative preferences in the proposed user profiles also help to separate similar and dissimilar queries into distant clusters, which helps to determine near optimal terminating points for our clustering algorithm.

Conclusion

We plan to take on the following two directions for future work.

1. First, relationships between users can be mined from the concept-based user profiles to perform collaborative filtering. This allows users with the same interests to share their profiles.

2. Second, the existing user profiles can be used to predict the intent of unseen queries, such that when a user submits a new query, personalization can benefit the unseen query.

Finally, the concept-based user profiles can be integrated into the ranking algorithms of a search engine so that search results can be ranked according to individual users’ interests.

Future Scope

Thank You