User Intent Discovery using Analysis of Browsing History

Abstract

The search engine can retrieve the information from the web by using keyword queries. The responsibility of search engines is getting the relevant results that met with users' search intents. Nowadays, all search engines provide search log of the user (queries logs, click information besides browsing history). The main objective of this work is to provide features that can help users during their web search by categorizing related browsing URLs together. That will be done by identifying intent groups for each URLs category, then identifying intent-segments for each intent group. Upon clustering the query categories, groups, and intent segments search engines can improve the representation of users' search context behind the current query, this would help search engines to discover the user's intents during the web search. Through the use of the normalized discounted cumulative gain (NDCG), the experimental results show the proposed method can improve the performance of the search engine.