Enabling Computer to Negotiate with Human in E-Commerce: A Strategy Model

Human-computer negotiation plays an important role in dynamic trading online, especially in B2C e-commerce. Further scientific investigations about designing the software agent that can deal with the human's random and inconsistent offer is in need, which is crucially useful for the online merchants to achieve better trading outcomes and save vast trading cost. The lack of such studies has decelerated the process of applying automated negotiation to real world applications. To address the critical issue, this paper develops a strategy model. To demonstrate the effectiveness of this model, we develop a prototype and conduct human-computer negotiation experiments over 121 participants. The experimental result shows that the agent with our newly designed strategy model can significantly increase the agreement rate and joint outcome of the both sides, and even can outperform human negotiators.