VISUAL ANALYSIS OF DISCUSSION FLOW ON KEYGRAPH-BASED BBS

Abstract

A method for visualizing discussion flow is proposed for analyzing the thread in KGBBS (KeyGraph-based BBS). It is known that group discussion is effective for chance discovery, in which information visualization plays an important role in providing participants with a material to that brainstorm various ideas, opinions, and interpretations. In order to enable online discussion while sharing the same visualized material, KGBBS is proposed based on KeyGraph, which is one of the typical visualization techniques in chance discovery. The aim of this paper is to utilize the result of discussion using KGBBS, by visualizing discussion flow of the thread in KGBBS. The method consists of 2 kinds of visualization techniques — a comment chain diagram to visualize the structure of comment chain, and a visual summary on KeyGraph to visualize concrete transition of topics through successive 2 or 3 comments. The proposed method is applied to actual discussion data, and the analysis results show the combination of both the visualization techniques, making it possible to analyze concrete transition of topics without reading comments in a thread.