Towards effective event detection, tracking and summarization on microblog data

摘要: Microblogging has become one of the most popular social Web applications in recent years. Posting short messages (i.e., a maximum 140 characters) to at any time and place lowers usage barrier, accelerates information diffusion process, makes it possible for instant publication. Among those daily userpublished posts, many are related or real-time events occurring our life. While microblog sites usually display list words representing trend topics during period (e.g., 24 hours, week even longer) on their homepages, topical do not make sense let users have comprehensive view topic, especially without background knowledge. Additionally, can only open each post relevant learn topic details. In this paper, we propose unified workflow event detection, tracking summarization data. Particularly, introduce novel features considering characteristics data selection, thus detection. phase, bipartite graph is constructed capture relationship between two adjacent time. The matched pair grouped into an chain. Furthermore, inspired by diversity theory search, first summarize chains content coverage evolution over experimental results show effectiveness approach