Fuzzy psychological behavior for computational bilateral negotiation

Having the best bilateral negotiation´s solution (win-win solutions for both parties) with incomplete information is a complex problem. Elsewhere internal agent´s character such as conciliatory or aggressive influences the attitude of cooperation and competitiveness, under external effects on negotiation process. The aim of this paper to develop a fuzzy psychological strategy for autonomous agent with incomplete information, where the characters conciliatory or aggressive, are suggested to be integrated in negotiation behaviors. For this purpose, first, one-to-one negotiation process, between a buyer agent with a fuzzy psychological behavior (our approach) and a seller agent with a time-dependent strategy (based on time dependent behaviors of Faratin et al.), over single issue (price), is exposed. Second, experimental environments and measures, allowing a set of experiments, carried out for different negotiation deadlines of buyer and seller agents, are detailed. Third, experimental results for both time-dependent and fuzzy psychological strategies are presented, analyzed, and compared for different deadlines of agents. The suggested approach allows agents to improve the negotiation process, with regard to the time dependent one, in terms of agent utilities, round number to reach an agreement, and percentage of agreements.