E²SGM: Event enrichment and summarization by graph model

In recent years, organizing social media by social event has drawn increasing attentions with the increasing amounts of rich-media content taken during an event. In this paper, we address the social event enrichment and summarization problem and propose a demon-

stration system E2SGM to summarize the event with relevant media selected from a large-scale user contributed media dataset. In the proposed method, the relevant candidate medias are rst retrieved by coarse search method. Then, a graph ranking algorithm is proposed to rank media items according to their relevance to the given event. Finally, the media items with high ranking scores are structured following a chronologically ordered layout and the textual metadata are extracted to generate the tag cloud. The work is concluded in an intuitive event summarization interface to help users grasp the essence of the event.