Applied graph mining technique to discover consensus graphs from group ranking decisions

摘要

The group ranking approach has been applied in many applications, such as in decision-making support systems, group recommendation systems, and so on. Previous studies have focused on how to generate a total ranking list. When there is no consensus or only slight consensus in the users' opinions or preferences, this kind of result may damage the decision-maker's decision. For this reason, this study proposes a new framework which represents user ranking information as a graph model, and a new algorithm based on the recent work to detect the maximum consensus graphs. The generated visualization results allow decision-makers to make better strategies.