AN ENHANCED TECHNIQUE FOR TEXT RETRIEVAL IN WEB SEARCH

S. Aarif Ahamed, V.Venkateshwaradevi, B.A.Vishnu Priya

ABSTRACT The web is a medium for accessing a great variety of information stored in various locations. As data on the web grows rapidly it may lead to several problems such as increased difficulty of finding relevant information. When a user submits a query to the search engine, it must be able to retrieve information according to the user s intention. But search engine retrieves the list of pages ranked based on their similarity to the query. Sometimes the results are not according to usersâ€™ interests, because many relevant terms may be absent from queries and words may be ambiguous. Therefore, the results produced by the search engine are not satisfactory to fulfill the user query request. In order to solve this ambiguity, the proposed system aim at discovering the number of diverse user search goals for a query and depicting each goal with some keywords automatically. First, the given user query is pre-processed to find the root word. Then, a feedback session has been introduced to infer user search goals for a query. Each feedback sessions are mapped to pseudo-documents using keywords which can efficiently reflect user information needs. Agglomerative Hierarchical clustering techniques is used to cluster the pseudo - documents. Improved HITS algorithm is used to rank cluster results relevant for a particular topic. The web results are restructured. Finally, we introduced user editable browser that allow the user to perform editing operations such as deletion and emphasis while browsing the search results. Index Terms :- user search goal, clicked URL, un clicked URL, VO â€“ pair, personalization, feedback session, pseudo â€“ documents, goal text, keyword, clustering, HITS, user editable browser.

Title:	AN ENHANCED TECHNIQUE FOR TEXT RETRIEVAL IN WEB SEARCH
Author Name:	S. Aarif Ahamed, V.Venkateshwaradevi, B.A.Vishnu Priya
Abstract:	ABSTRACT The web is a medium for accessing a great variety of information stored in various locations. As data on the web grows rapidly it may lead to several problems such as increased difficulty of finding relevant information. When a user submits a query to the search engine, it must be able to retrieve information according to the user s intention. But search engine retrieves the list of pages ranked based on their similarity to the query. Sometimes the results are not according to usersâ€™ interests, because many relevant terms may be absent from queries and words may be ambiguous. Therefore, the results produced by the search engine are not satisfactory to fulfill the user query request. In order to solve this ambiguity, the proposed system aim at discovering the number of diverse user search goals for a query and depicting each goal with some keywords automatically. First, the given user query is pre-processed to find the root word. Then, a feedback session has been introduced to infer user search goals for a query. Each feedback sessions are mapped to pseudo-documents using keywords which can efficiently reflect user information needs. Agglomerative Hierarchical clustering techniques is used to cluster the pseudo - documents. Improved HITS algorithm is used to rank cluster results relevant for a particular topic. The web results are restructured. Finally, we introduced user editable browser that allow the user to perform editing operations such as deletion and emphasis while browsing the search results. Index Terms :- user search goal, clicked URL, un clicked URL, VO â€“ pair, personalization, feedback session, pseudo â€“ documents, goal text, keyword, clustering, HITS, user editable browser.
Cite this article:	S. Aarif Ahamed, V.Venkateshwaradevi, B.A.Vishnu Priya , " AN ENHANCED TECHNIQUE FOR TEXT RETRIEVAL IN WEB SEARCH " , International Journal of Emerging Trends & Technology in Computer Science (IJETTCS), Volume 4, Issue 1, January - February 2015 , pp. 006-010 , ISSN 2278-6856.
Full Text [PDF] Home