Cognitive Fuzzy-based Behavioral Learning System for Augmenting the Automated Multi-issue Negotiation in the E-commerce Applications

Abstract

Evolution of agent-based technology presents behavioral learning and sustainable negotiation challenges in e-commerce applications. In particular, the challenge of designing the negotiation strategy to incorporate sustainability in e-commerce business that can leverage the agent to reach its objectives by increasing the negotiation coordination and cooperation with the opponent agents. Therefore, the proposed research introduces the negotiation strategy sustainable solution using a cognitive fuzzy-based behavioral learning system which can change the preferences of negotiating agents according to human psychological characteristics. It will mimic the attitudes of human risk, patience and regret during the course of bilateral negotiation and also change the preference structures according to the fuzzy logic rules. As a result, the proposed negotiation strategy makes significant improvements on various parameters such as utility value, success rate, total negotiation time, and communication overhead while changing the negotiation rounds from 50 to 500. Since this system leverages the negotiation strategy of the agent by taking appropriate decisions to reach better agreement based on the interest, belief and psychological characteristics of negotiating opponents. Moreover, the usage of negotiation in the cloud-based platform can leverage the e-commerce applications to handle as many requests as possible due to its dynamic elasticity.