From Tweet to Graph: Social Network Analysis for Semantic Information Extraction

Rocio Abascal-Mena (1) , Rose Lema (1) , Florence Sèdes (2)

1 UAM - Universidad Autónoma Metropolitana [Mexico]
2 IRIT-SIG - Systèmes d’Informations Généralisées

This paper represents a study along the cutting edge of the current analysis of online social network in relation with the contents communicated among users. Twitter data is carefully selected around a fixed hash-tag in order to study the specified content in relation with other contents that users bring to connection. A separate network of hash-tags related (in tweets) is constructed for different days; the networks are analyzed within advanced Gephi package, providing several measures -degree, betweenness centrality, communities, as well as the longest path, by which the evolution of communication around specified concepts is quantified. Our study is absolutely in the current trend of analysis of online social networks that, going beyond mere topology, reveals relevant linguistic and social categories and their dynamics.

Mots clés: Social Network Analysis; Community detection; Twitter; Theory of Graphs; Social Web; Text mining

Domaines: Théorie de l'information [cs.IT]; Recherche d'information [cs.IR]

Rocio Abascal-Mena, Rose Lema, Florence Sèdes. From Tweet to Graph: Social Network Analysis for Semantic Information Extraction. IEEE International Conference on Research Challenges in Information Science (RCIS 2014), May 2014, Marrakesh, Morocco. pp.1-10, ⟨10.1109/RCIS.2014.6861047⟩. ⟨hal-01147320⟩