Weaving the Topics of CHI: Using Citation Network Analysis to Explore Emerging Trends

This paper provides a comprehensive and novel analysis of the annual conference proceedings of CHI to explore the structure and evolution of the community. Self-awareness is healthy for a diverse and dynamic community, allowing it to anticipate and respond to emerging themes. Instead of using a traditional topic modelling approach to analyze the text of the papers, we adopt a social network analysis approach by analyzing the citation network of papers. After constructing such a citation network, community detection is applied in order to cluster papers into different groups. The keywords of these groups are found to represent different research themes in human-computer interaction, while the growth or decline of these groups is visualized by their paper shares over time. Lastly, we contribute a visualization tool for exploring emerging trends within our community, which can be used to predict likely topics at future CHI conferences.