Estimation of the buyer´s contract space incorporating learning from experience techniques to the seller´s rationale in e-commerce context

Various state-of-the-art technologies are necessary to enhance the efficiency and increase the interest for e-commerce transactions. Mobile agents are one of the means that may enhance the intelligence and improve the effectiveness of systems in the e-marketplace. This paper aims to present the basic elements of the designed dynamic multilateral negotiation model and strategies that do not require a complicated rationale on behalf of the buyer agents. It focuses on the enhancement of the seller´s reasoning component by incorporating to the designed negotiation strategies a novel mechanism for the estimation of the mutually acceptable contract region by exploiting relative market data combined with knowledge acquired from previous experience. This technique is used to extend the functionality of autonomous agents, so that they reach to an agreement faster aiming to maximise their owner´s utility. The framework considers both contract and decision issues, is based on real market conditions, and has been empirically evaluated.