Multi-strategy Selection Model for Automated Negotiation

Automated negotiation has played an important role in supporting the dynamic trading based e-commerce. Research in automated negotiation, especially for computer-computer negotiation pays little attention on the implementation related issues such as multi-strategy selection, which will be very useful for the computer-human negotiation. The strategy selection is very important for negotiating agent to achieve better negotiation outcomes. The lack of such study has slowed down the process of applying automated negotiation to real world problems. To address the issue, this paper develops a multi-strategy selection model grounded on the integration of the time-dependent and behavior-dependent tactics, the multi-strategy selection theoretical model and algorithm is proposed to meet the following three goals: supporting multi-strategy selection, easy to be implemented and less resources consuming. To demonstrate the effectiveness of the proposed model and algorithm, a prototype of the model is built, and a lot of experiments are conducted to demonstrate the effectiveness of the model.