A Statistical Model for User Preference
IEEE Transactions on Knowledge and Data Engineering
Resume information extraction with cascaded hybrid model
ACL '05 Proceedings of the 43rd Annual Meeting on Association for Computational Linguistics
The dynamics of interactive information retrieval behavior, Part I: An activity theory perspective
Journal of the American Society for Information Science and Technology
Evaluating Bayesian networks' precision for detecting students' learning styles
Computers & Education
Learning drifting concepts: Example selection vs. example weighting
Intelligent Data Analysis
A Combination Approach to Web User Profiling
ACM Transactions on Knowledge Discovery from Data (TKDD)
Empirical analysis of predictive algorithms for collaborative filtering
UAI'98 Proceedings of the Fourteenth conference on Uncertainty in artificial intelligence
Hi-index | 0.00 |
With rapid development of Internet, information and resource of Web academic database is great and explosive growth, so it is difficult to quickly and accurately obtain information which meets individual user's needs. Web personalized services can effectively solve the problem of information overload problem and alleviate user's cognitive burden. How to predict user interest is a key issue in Web personalized services. First, this paper proposes concepts of user knowledge unit and user knowledge flow that represents user short-term interest and long-term interest respectively. Second, existing methods have some defects which can't be sensitive to perceive user interest change and accurately predict user real-time interest; we put forward Collaborative Time Weight (CTW) and Collaborative Relation Weight (CRW) to solve those problems. Meanwhile the prediction algorithm for user real-time interest is proposed based on collaborative filtering and interactive computing. Finally, experimental results demonstrate that our method can capture user real-time interests accurately and alleviate the user's cognitive burden effectively.