Abstract: Agent persuasion of automation negotiation can improve the efficiency of negotiation enough at dynamic environment，which can guarantee its effect and save its cost greatly.Case-Based Reasoning（CBR） can make agent comparing the new case that meets in negotiation with the successful case in its case database，then make the better selection according with it，which can bring multi-agent to cooperate well.The definition of matching degree of case in agent persuasion is given.Based on it，relative model is constructed with formal logic theory，which is illustrated by an example.The method of calculation about key index in this model is given，and relevant evaluation system and interactive mechanism is also proposed.Finally agent persuasion mechanism based on its matching degree of case is built and verified by another example.Results show that this mechanism can decrease the cost of this system effectively.

Key words: persuasion mechanism, matching degree, case

摘要： 自动谈判中的Agent劝说能充分提高其在动态环境下的谈判效率，保证谈判效果，同时极大节约谈判成本。案例推理技术能使Agent将谈判中遇到的新案例与以往成功案例进行比对，根据结果做出较优的选择，从而促使Agent之间取得更佳的合作效果。对Agent劝说中的案例匹配度的概念进行了界定；结合形式逻辑理论，建立了基于案例匹配度的Agent劝说模型，并通过举例对模型进行了阐述；提出了模型中关键指标的计算方法，并通过算例进一步给出了相应的评价体系和交互机制；最后进行了验证，实现了基于案例匹配度的Agent劝说机制的构建。实验结果表明，该机制能有效减少系统开销。

关键词: 劝说机制, 匹配度, 案例