Community topical fingerprint analysis based on social semantic networks

Abstract

Community analysis of social networks is a widely used technique in many fields. There have been many studies on community detection where the detected communities are attached to a single topic. However, an overall topical analysis for a community is required since community members are often concerned with multiple topics. In this paper, we propose a semantic method to analyze the topical community fingerprint in a social network. We represent the social network data as an ontology, and integrate with two other ontologies, creating a Social Semantic Network (SSN) context. Then, we take advantage of previous topological algorithms to detect the communities and retrieve the topical fingerprint using SPARQL. We extract about 210,000 Twitter profiles, detect the communities, and demonstrate the topical fingerprint. It shows human-friendly as well as machine-readable results, which can benefit us when retrieving and analyzing communities according to their interest degrees in various domains. © Springer Science+Business Media Dordrecht 2014.