A novel approach for modeling users short-term interests, based on user queries
In this paper is presented a novel approach for modeling user short-term interests, based on user queries. The proposed user model is represented by a weighted semantic network, composed of nodes and arcs. This semantic network can be used to express relations between query terms submitted by a user in his searches. Our approach is rooted on the idea that there are relations between topics of users interests, which can be measured and used to provide a search context. We propose an approach for modeling users interests, based on data taken from his previous queries. We aim to identify relations between topics of users interests in a set of successive queries. The search personalization features of ShareTec digital library are also presented and the implementation of the proposed model is outlined.
Keywords: user profile, short-term interests, adaptive query, search personalization
Download Full-Text
ABOUT THE AUTHOR
Albena Turnina
PhD Student
Albena Turnina
PhD Student