Micro-blog topic recommendation based on knowledge flow and user selection

Micro-blog topic recommendation aims to solve the problem of low efficiency for micro-blog topic recommendation caused by excessive micro-blog data. This paper proposed a micro-blog topic recommendation based on knowledge flow and user selection to improve the accessing speed of micro-blog and efficiency of topic recommendation. The micro-blog topic recommendation's core tasks have two sides. One is analyzing the user's preference for the micro-blog topic based on the user's historical behavior. The other is recommending the topic to other users who have the similar historical behavior. First, users are clustered according to users' previous preference to micro-blog topic. After that, the micro-blog topics of knowledge flow in different class (i.e., belongs to different users) are recommended. Finally, the knowledge flow according to the user selection of recommended topics is updated to improve the accuracy of micro-blog topic recommendation. The experimental results show that the proposed algorithm can improve the accuracy and efficiency of micro-blog topic recommendation effectively.