Mining of Social Networks from Classic Books and Visualization

摘要

While modern online social networks offer unprecedented amount of data and opportunities for mining, we acknowledge that other forms of social networks have been around throughout human history. In particular, we recognize the classic books as valuable sources for the analysis of interpersonal interactions as they have had positive influence on human mind and continue to enlighten the human race. In this paper, we propose a formal framework for the mining of interpersonal interactions from classic books. Our approach employs a heuristic over a traditional named-entity recognition for extracting people names. We demonstrate the effectiveness of our approach by presenting the encouraging result from a network analysis conducted over the persons appearing in the book of Genesis. The result aids the understanding of the persons featured in the book and their interactions. We also present the merits and challenges of visualization techniques pertaining to our work.