Abstract:

Topic in microblog data is dynamic. Aiming at detecting microblog’s topic more precisely, we should make use of the context of given microblog other than its text content. This paper proposes a temporal related topic detection method. This method can be divided into four phases: preprocessing, feature selection, topic detection, topic evolution analysis. Experimental results on real data show that the methods we propose in this paper perform well.

Key words: Topic detection, 　Microblog, 　Feature Selection, 　Topic Model