Event Details

Citation network analysis of a computer science conference (Clement Lee)

Understanding and discovering the themes of an academic field is usually done by analysing the text of a corpus of publications. In this work we adopt an alternative approach, by constructing a citation network using the references of the publications, and fitting a network clustering model. This model is being applied to the ~6000 papers of a computer science conference. Post-clustering text analysis reveals that the different groups of papers match well with the major themes of the conference.