Fuzzy Driven Multi-issue Agent Negotiation on Electronic Marketplace

Agent-based online negotiation by different autonomous agents has become radically important since the advent of e-business. Due to recent growing interest in the electronic marketplace, scientists are developing numerous models of inter-organizational electronic commerce. Most of the negotiation processes are complicated and time consuming. In this paper, we develop an intelligent negotiation model based on fuzzy constraints for bilateral multi-issue negotiation in business environments. The proposed model consists of two agents. The Information Agent (IA) is trained by the Knowledge Base (KB) and Negotiator Agent (NA) caries out the negotiation process. The aim of the NA in this model is to reach mutual agreement effectively and intelligently, while both agents are able to learn from the environment. The negotiation process is driven by fuzzy logic, considering the agent's satisfaction within e-commerce system framework. In this paper, the system uses the KB before starting the negotiation process. Additionally, the KB is used, in different negotiation stages, to train fuzzy constraints for carrying out negotiations. We propose an equation that incorporates qualitative and quantitative characteristics to generate the next offering price in the negotiation. The Fuzzy Rule base is responsible for generating a rational value. This rational value, based on fuzzy constraints, controls the agent's satisfaction level.