Understanding support of group in web collaborative learning, based on divergence among different answering processes

Abstract

Recently, web based collaborative learning is one of the hottest subject. In collaborative learning, not only deriving answer, but also considering various methods is important in order to get the ability to deal with different exercises. Therefore, our objective is to grasp the process of solving answer and specify the derived answering paths, then generate advice to make the learning effectual. In this paper, we focus on the mechanism to grasp the extent of discussion and detect answering methods that students need to discuss about. Our approach is to arrange the answering paths on “discussion-extent projection” based on the similarity among them and detect the answering paths that include different viewpoint than derived one. In addition, we address “divergent tree” which represents whole answering paths with respect to the divergent points among them in order to calculate the similarity.