Visualizing Topics and Social Networks within a Health Newsgroup

Introduction The goal of this project was to develop a visualization that would simultaneously overlay topics extracted from a newsgroup with social networking information. The motivation of this work is to determine if social networking information can be used as a method to evaluate the quality of topic extraction from a newsgroup. This work is also motivated by the questions: can hierarchical clustering and summarization of newsgroup messages lead to insights about the formation and sustainability of social networks? How can individual analysis be combined to inform another analysis? Ideally, a synergistic effect would be discovered, leading to insights about the community as a whole.