Large-scale content analysis, Natural language processing
National Center for Scientific Research (NCSR) “Demokritos”
Title of Paper:
Social Web Observatory: A Platform for Gathering Knowledge from the Social Web
Subdisciplinary Area: AI and Social Web
Keywords: Social Web Observatory, information diffusion, listen-search-analyze-analytics, infrastructure
Abstract: The Social Web Observatory is an initiative aiming to help researchers (mainly of the social sciences and digital humanities) to study information diffusion in the social web (news and user generated content - such as comments and posts in social media networks). The Social Web Observatory integrates the following subsystems: a) Listen: The Social Web Observatory listens a wide variety of news sources and user generated content (such as comments in DISQUS and tweets in Twitter); b) Search: Content is indexed through a search infrastructure, enabling users to retrieve context through sets of keywords, for further analysis; c) Analyse: Content retrieved through search is analysed along various dimensions to extract indicators such as trends, coverage, events, sentiment, etc.; d) Analytics: Both context and indicators are visualised through dashboards and other analytics tools, to provide information and insights on the various issues. The infrastructure allows the user to take advantage of the gathered information as actionable knowledge: to understand the dynamics of the public opinion for a given entity over time and across real-world events. In this paper we describe the platform, the analysis functionality and we evaluate the performance of the system on a specific use case.