Multiagent Decision Support Systems based on supervised learning

Decision support systems (DSS) are a specific class of computerized information systems that supports business and organizational decision-making activities. Nowadays, agent oriented techniques are important new means in analyzing, designing and building complex software systems. We aim in this paper to present a generic architecture for a multiagent decision support system that uses supervised learning techniques for assisting users in their decision making tasks. We have emphasized the advantages of the proposed architecture and we have implemented, based on the proposed architecture, a multiagent system for supporting physicians in their diagnosis tasks. The performance of our system and comparisons with other similar approaches are also reported.