标题：A dynamic filtering recommendation algorithm based on topic
作者：An, Xinjun ;Su, Na
作者机构：[An, Xinjun ;Su, Na ] Department of Information Engineering, ShanDong University of Science and Technology, Tai'an, ShanDong, China
会议名称：2011 International Conference on Electronic and Mechanical Engineering and Information Technology, EMEIT 2011
会议日期：August 12, 2011 - August 14, 2011
来源：Proceedings of 2011 International Conference on Electronic and Mechanical Engineering and Information Technology, EMEIT 2011
摘要：Collaborative filtering is one of the most successful and widely used recommendation technology in E-commerce recommendation systems. However, existing collaborative filtering algorithms face severe challenge of sparse user ratings and real-time recommendation. To solve the problems, a collaborative filtering recommendation algorithm based on topic is proposed. It divides the raw rating matrix into many sub-matrixes based on topic and forms clusters parallelly to reduce the data sparsity. Time-based data weight and acceleration-based data weight are proposed to dynamically reflect the change of user interests. The experimental results show that the novel algorithm can efficiently improve recommendation quality. © 2011 IEEE.