Mining Social Response to Crisis via Electronic Media

In this paper, we are interested in using text mining to analyze textual data obtained from electronic social media in order to find out social concerns on crisis. In particular, we explore the temporal theme patterns in the natural catastrophe, i.e. the Yushu Earthquake, and compare the foci of the Chinese public, Chinese government and the overseas. By incorporating sources of information with a variety of aspects, we can distill not only snapshots of social concerns at given times but also the evolution of these concerns over time. Our intension is to assist the government in publicizing appropriate information and making intelligent decisions.