Mining collaboration through textual semantic interpretation

Saurav Karmakar, Ying Zhu

Abstract: In this paper, we report our attempt to evaluate the degree of online collaboration with quantitative analysis. Since most online collaborations are carried out through natural language, such quantitative analysis must rely on natural language processing. Here we discuss our work in analyzing an online social visualization web site using the semantic similarity measures of collaborative textual snippets. We use word-based textual similarity measures to study the relationships among user comments. We also detect and visualize the patterns of user collaborations.