PREFERENCE-BASED DECISION-MAKING FOR COOPERATIVE KNOWLEDGE-BASED SYSTEMS

Recent advances in cooperative knowledge-based systems (CKBS) offer significant promise for intelligent interaction between multiple AI systems for solving larger, more complex problems. In this paper, we propose a logical, qualitative problem-solving scheme for CKBS that uses social choice theory as a formal basis for making joint decisions and promoting conflict resolution. This scheme consists of three steps: (1) the selection of decision criteria and competing alternatives, (2) the formation of perference profiles and collective choices, and (3) the negotiation among agents as conflicts arise in group decision making. In this paper, we focus on the computational mechanisms developed to support steps (2) and (3) of the scheme. In addition, the practicality of the scheme is illustrated with examples taken from a working prototype dealing with collaborative structural design of buildings.