Learning Mechanism of Automated Negotiation in E-commerce

Aiming at the shortcoming of current automated negotiation systems, this paper applied machine learning to bilateral automated negotiation. It mainly researched learning mechanism of automated negotiation in e-commerce. It improved traditional Q-learning and designed dynamic Q-learning algorithm. This algorithm estimated Q value according to environment state and the action of both agents, furthermore, recency-based exploration bonus were embedded. The paper applied Bayesian learning to negotiation strategy of automated negotiation, and designed the belief strategy based on Bayesian. Finally, this paper did experiments on learning mechanism and dynamic Q-learning algorithm. The results show that learning mechanism can improve efficiency of automated negotiation and dynamic Q-learning algorithm is efficient