Event Detection: Monitoring and Tracking the Dynamics of Social Networks Communities

Abstract

This paper presents research done towards a robust real-time social network text stream event detection system that combines text stream mining and network analysis methods. It presents the current state-of-the-art systems, algorithms, and methodologies to perform event detection in streaming environments: If from the point of view of a natural language processing, text mining, and unsupervised learning the problem of detecting events in unbounded text streams is hard, dealing with dynamic networks with millions of nodes and edges is not also an easy task. Presented contributions and research directions are based on the premise that the precision and accuracy of an event detection algorithm could be improved by considering network properties of the social network when events happen. Network analysis algorithms, specifically algorithms to perform dynamic community detection, community identification and community tracking can be used to extract knowledge from users relations and interactions helping in the task of unveiling and detecting new or unforeseen events.