Including Social Factors in an Argumentative Model for Group Decision Support Systems

Abstract

In this paper we propose a decision support system for groups of people where each user delegates to an agent that represents her preferences and argues with other agents to obtain the best alternative for the whole group. The novelties of our approach are the inclusion of users’ social factors, personality and trust, in the argumentation process and the negotiation system, plus a multi-agent architecture that represents the social connections within the group. Therefore, our model simulates the argumentations made by real users to agree on a concrete product in a very accurate way. As a case study, we have tested our theories in the movie recommendation domain with real social networks. We have concluded that distributed models and argumentation techniques including personality and social trust improve the satisfaction of users involved in a group decision making process.