Award Date

12-2011

Degree Type

Thesis

Degree Name

Master of Science in Computer Science

Department

Computer Science

First Committee Member

Kazem Taghva, Chair

Second Committee Member

Ajoy Datta

Third Committee Member

Laxmi Gewali

Graduate Faculty Representative

Venkatesan Muthukumar

Number of Pages

74

Abstract

A typical day of million web users all over the world starts with a simple query. The quest for information on a particular topic drives them to search for it, and in the pursuit of their info the terms they supply for queries varies from person to person depending on the knowledge they have. With a vast collection of documents available on the web universe it is the onus of the retrieval system to return only those documents that are relevant and satisfy the user’s search requirements. The document mismatch problem is resolved by appending extra query terms to the original query which improves the retrieval performance. The addition of terms tends to minimize the bridging-gap between the documents and queries.

In this thesis, a brief study is done on the reformulation of queries, along with methods of calculating the relevancy of candidate terms for query expansion by using several ranking algorithms, term weighting algorithms and feedback processes involving evaluations. Comparisons of various methods based on their efficiencies are also discussed. On the whole a consolidated report of query expansion in general is given.

Keywords

Applied sciences; Efficient expansion; Feedback evaluation; Query expansion; Querying (Computer science); Ranking algorithms; Relevance feedback; Search engines – Programming; Term weighting; Web search engines; Web searching

Disciplines

Computer Sciences | Databases and Information Systems | Systems Architecture

Language

English


Share

COinS