A Model for Summarizing Celebrities with Microblogging Users' Interest

With the rapid development of Web 2.0, microblogging such as twitter is increasingly becoming an important source of up-to-date topics about what is happening in the world. Specially, it becomes one of the main ways for users to understand some celebrities. However, the huge volume of information makes troubles for people to get what they really want. How to filter out needless information through numerous microblogging data and form a brief review of a celebrity become necessary for microblogging users. In this paper, we propose a novel solution for understanding a celebrity by summarizing microblogging users’ interest on him/her, and a framework is outlined. Users’ interest are generally shown on the most salient historical events of the given celebrity, and usually users’ emotion can be reflected with their microblogs contents. Therefore, the proposed model first utilizes a burst detection algorithm to extract burst periods of the time series of given celebrity’s microblogs, and then conduct emotion analysis on the microblogs in burst event periods. Experimental results show that the proposed solution can effectively summarize celebrities with microblogging users' interest. Keywords—Microblogging; text summarization; celebrity summarization; user interest; burst detection; user emotion analysis.