Towards enhancing historical analogy: Clustering users having different aspects of events

Abstract

Studying history can provide numerous benefits for finding meaningful connections or analogies over time. Several researchers have studied how to support promoting historical analogy by computer supported learning; however, supporting group discussions to promote the analogy still remains unexplored. In this paper, we propose a novel clustering algorithm to form users who have different aspects of the same past event in order to ease exchange ideas of what aspects they focus to analyze the event. We implemented our algorithms and evaluated them in terms of getting accuracies of forming users having different aspects. Experimental results proved that only our algorithm creates suitable groups.