Dynamic topic mining for microblog fused with user's behavior and time window

Compared with traditional text, microblog text has features of user behavior and time window. Catered to features of microblog text, this paper proposed a method of dynamic topic mining for Microblog fused with user behavior and time window. Based on traditional LDA model, we use method of time window division to divide microblog text into each time window, then fuse features of user behavior in our model as guide information, sequentially construct dynamic topic mining for Microblog fused with user behavior and time window. Result of the experiment shows that the model we proposed has better effect on topic in microblog analyzing and topic intensity changing with time.

[1]  Jon Whittle,et al.  A Feasibility Study on Extracting Twitter Users' Interests Using NLP Tools for Serendipitous Connections , 2011, 2011 IEEE Third Int'l Conference on Privacy, Security, Risk and Trust and 2011 IEEE Third Int'l Conference on Social Computing.

[2]  Kristina Lerman,et al.  Analyzing microblogs with affinity propagation , 2010, SOMA '10.

[3]  Mark Steyvers,et al.  Finding scientific topics , 2004, Proceedings of the National Academy of Sciences of the United States of America.

[4]  Myra Spiliopoulou,et al.  Topic Evolution in a Stream of Documents , 2009, SDM.

[5]  Fabio Crestani,et al.  Evaluation of an interactive topic detection and tracking interface , 2012, J. Inf. Sci..

[6]  Cem Aksoy,et al.  Novelty detection for topic tracking , 2012, J. Assoc. Inf. Sci. Technol..

[7]  Susan T. Dumais,et al.  Characterizing Microblogs with Topic Models , 2010, ICWSM.