| 基于偏好的双边多议题协商优化 |
| |
| 引用本文： | 王永强,刘胜全,曹冠平.基于偏好的双边多议题协商优化[J].计算机应用与软件,2012(3):175-178. |
| |
| 作者姓名： | 王永强 刘胜全 曹冠平 |
| |
| 作者单位： | 新疆大学信息科学与工程学院;新疆大学现代教育技术中心 |
| |
| 基金项目： | 新疆维吾尔自治区科技攻关项目(200931103) |
| |
| 摘 要： | 提出一种优化的自动协商模型。Agent在信知不完全的情况下通过学习交互历史和在线协商信息获取对手的偏好,结合贝叶斯方法和支持向量机学习对手偏好,基于保留值和权重提出一种决策模型。通过实验比较和分析,该模型能有效降低协商次数,提高协商双方的联合效用。在信息保密和先验知识未知的环境下,该模型仍然表现出了较高的效用和效率。
|
| 关 键 词： | 自治协商 偏好 协商策略 贝叶斯学习 JADE |
PREFERENCE-BASED BILATERAL MULTI-ISSUE NEGOTIATION OPTIMIZATION |
| |
| Wang Yongqiang,Liu Shengquan,Cao Guanping.PREFERENCE-BASED BILATERAL MULTI-ISSUE NEGOTIATION OPTIMIZATION[J].Computer Applications and Software,2012(3):175-178. |
| |
| Authors: | Wang Yongqiang Liu Shengquan Cao Guanping |
| |
| Affiliation: | 1(School of Information Science and Engineering,Xinjiang University,Urumqi 830046,Xinjiang,China) 2(Modern Education Technology Center,Xinjiang University,Urumqi 830046,Xinjiang,China) |
| |
| Abstract: | This paper proposes an optimal autonomous negotiation model with which the agent can obtain the opponent’s preference information through learning interactive history and online consultations information when information is acquired incompletely;then combining Bayesian methods and support vector machine it can learn component’s preference,and a decision model which is based on reservation value and weight preferences is proposed.Comparison and analysis of experiments prove the model can effectively reduce the number of consultations and improve the joint utility of both consultative parties.Under such environments as secret information and unknown prior knowledge,the model still behaves high effectively and efficiently. |
| |
| Keywords: | Automated negotiation Preference Negotiation strategy Bayesian learning JADE |
| 本文献已被 CNKI 等数据库收录！ |
|
