An intelligent idea categorizer for electronic meeting systems

Abstract

Research on group decisions and electronic meeting systems have been increasing rapidly according to the widespread of Internet technology. Although various issues have been raised in empirical research, we will try to solve an issue on idea categorizing in the group decision making process of electronic meeting systems. Idea categorizing used at existing GDSS was performed in a top-down procedure and mostly by participants manual work. This resulted in tacking as long in idea categorizing as it does for idea generating, clustering an idea in multiple categories, and identifying almost similar redundant categories. However such methods have critical limitation in the electronic meeting systems, so we suggest an intelligent idea categorizing methodology which is a bottom-up approach. This method consists of steps to present idea using keywords, identifying keywords affinity, computing similarity among ideas, and clustering ideas. This methodology allows participants to interact iteratively for clear manifestation of ambiguous ideas. We also developed a prototype system, IIC (Intelligent Idea Categorizer) and evaluated its performance using the comparison experiments with other systems. IIC is not a general purposed system, but it produces a good result in a given specific domain.