Discovering knowledge in Microblog based on naturally annotated web resources

Microblog has become one of the most popular social networking services. Users of Microblog generate large amounts of data everyday. Volumes of data and the large number of users make Microblog potentially a valuable data source for data mining and knowledge discovery. This paper aims to mine knowledge from the user generated data and discover the major knowledge contributors in Microblog. We propose a framework based on naturally annotated web resources and apply this framework to mine three kinds of knowledge and to discover the major knowledge contributors in Microblog. Experimental results show the effectiveness of the proposed framework.