Empirical study of an intelligent argumentation system in MCDM

Abstract

Intelligent argumentation based collaborative decision making system assists stakeholders in a decision making group to assess various alternatives under different criterion based on the argumentation. A performance score of each alternative under every criterion in Multi-Criteria Decision Making (MCDM) is represented in a decision matrix and it denotes satisfaction of the criteria by that alternative. The process of determining the performance scores of alternatives in a decision matrix for criterion could be controversial sometimes because of the subjective nature of criterion. We developed a framework for acquiring performance scores in a decision matrix for multi-criteria decision making using an intelligent argumentation and collaborative decision support system we developed in the past [1]. To validate the framework empirically, we have conducted a study in a group of stakeholders by providing them an access to use the intelligent argumentation based collaborative decision making tool over the Web. The objectives of the study are: 1) to validate the intelligent argumentation system for deriving performance scores in multi-criteria decision making, and 2) to validate the overall effectiveness of the intelligent argumentation system in capturing rationale of stakeholders. The results of the empirical study are analyzed in depth and they show that the system is effective in terms of collaborative decision support and rationale capturing. In this paper, we present how the study was carried out and its empirical results.