A Personalized Result Recommendation Method based on Communities
Ying-Jian Li, Qingshan Li, Yishuai Lin · 2017
The vagueness of queries is a thorny problem of web search all the time. Because of the length of queries is general short, identifying user's search intent from a single query session is unpractical. In this paper, a result recommendation method which refers to the idea of swarm intelligent to optimizing the result list is proposed to solve this problem. In this method, users will be allocated to several communities where users have similar features of searching or other characteristics. And users will benefit from the search experience of co-community users when they search the repetitive queries. Furthermore, to optimizing the result list, the most relative results from similar users will be place ahead of the result page. Finally, experimental evaluation shows that our approach improves the accuracy of results and personalized user experience of searching.