Intelligent computational argumentation for evaluating performance scores in Multi-Criteria decision making

Multi-Criteria decision making (MCDM), is a discipline aimed at assisting multiple stakeholders in contemplating a decision paradigm in an uncertain environment. The decision analysis to be performed involves numerous alternative positions assessed under varied criterion. A performance score is assigned for each alternative in terms of every criterion and it represents satisfaction of the criteria by that alternative. In real applications, performance scores are sometimes hard to determine and they are often subjective. We have developed an intelligent computational argumentation approach for dealing with the problem of uncertainty in resolving the subjective scores. In this approach, an argumentation tree is developed to assess a performance score for an alternative under a criterion. The argumentation takes into consideration the strength of an argument i.e. the degree of support or attack for that argument, and priorities of stake-holders. A set of fuzzy argumentation rules in a fuzzy association matrix is used to assess the indirect impact of an argument on alternatives. Aggregation of strengths of supporting and attacking; direct and indirect arguments represents a performance score of an alternative for a criterion in the decision making domain. A decision making case study for developing a mine detection simulator is used to illustrate the method.