An Argumentative Knowledge-Based Model Construction Approach for Bayesian Networks

Abstract. In this paper, an argumentative knowledge-based model con-struction (KBMC) technique for Bayesian networks is presented. This ap-proach allows an agent to collect and instantiate the most accepted subset of an imperfect knowledge base to dynamically construct a Bayesian net-work. Arguments are constructed to represent active paths through an agent’s knowledge base – paths consisting of information that is compu-tationally relevant in the evaluation of a query Pr(Q|E). Argumentation over paths is used to select the valid or most accepted information ac-cording to the preferences of the agent. This information is consequently formed into candidate network structures by accrual. This work is pre-sented as an extension of the KBMC approach of Haddawy [5]. The po-tential of the approach to be used in multi-agent network construction is discussed.