Study on multi-agent based simulation process of signaling game in e-commerce

Although customer purchase behaviors are usually affected by the intrinsic qualities of the product, so are the influence of its social network and business behaviors. By applying signaling game theory and multi-agent modeling and simulation method, the paper focuses on the customer behaviors and the transaction process with consideration of the learning from customer's social network and the impact of the business behavior. First, the process of signals transferred between business and customer is analyzed. Then, an anticipated product quality model based on its experience and its social network's experience is given where quality uncertainty exists. Moreover, the utilities of both business and customer are established. Finally, a simulation process of signaling game between business and customer based on multi-agent is designed and a prototype of simulation is build, in order to offer a decision support for the transaction under asymmetric information.