A computational model for multi-agent E-commerce negotiations with adaptive negotiation behaviors

This paper presents a computational model to organize multi-agent E-commerce negotiations with adaptive negotiation behaviors aiming at enhancing the negotiation flexibilities of software agents. Firstly, the computational E-commerce negotiation model covering negotiation protocol, negotiation issues and negotiation strategies is specified to assist agents’ computing functions. Then, a three-staged adaptive negotiation behavior configuration mechanism is proposed to tackle the negotiation dynamics. In the pre-negotiation stage, agents’ negotiation behaviors are deployed by the case-based strategy assignment mechanism; in the on-going negotiation stage, opponents’ negotiation behaviors are tracked through the neural network learning model; in the post-negotiation stage, opponents’ concession functions are recorded using the time series measure. Finally, the computational negotiation model is tested through hypothetical negotiation cases. The outcomes show that the adaptive negotiation behavior configuration mechanism can benefit an agent to win more in the E-commerce negotiation.