Making Sense of Micro-posts for Organizations and Businesses: Live Event and User Community Detection

Abstract

As a massive repository of User Generated Content (UGC), social media platforms are arguably the most active networks of interactions, content sharing, and news propagation that best represent the everyday thoughts, opinions and experiences of their users. Rapid analysis of such contents is thus critical for user-centric organizations and businesses as the relevant social media contents provide actionable insights for such organizations. This thesis focuses on online discovery and analysis of the social media contents for organizations. We propose algorithms to effectively harvest relevant data about a given organization from social media, identify the emerging and evolving discussions about the organization as well as its user community. Our mining algorithms utilize information about current keywords, users, micro-posts, topics, and opinions about organizations to tackle the above issues. Extensive experiments on different kind of UGCs show the effectiveness of the proposed approaches.