中山大学学报自然科学版 ›› 2011, Vol. 50 ›› Issue (1): 18-22.

• 研究论文 • 上一篇    下一篇

一种竞争环境下基于改进遗传算法的电子商务协商

李 剑1,2,景 博2,袁彩霞1
  

  1. (1.北京邮电大学网络与交换国家重点实验室,北京 100876;2.北京理工大学软件学院,北京 100081)
  • 收稿日期:2010-01-09 修回日期:1900-01-01 出版日期:2011-01-25 发布日期:2011-01-25

A Negotiation Based on Improved Genetic Algorithm and Its Application in Competitive E-commerce

LI Jian 1,2, JING Bo 2, YUAN Caixia 1
  

  1. (1. State key Lab of Network and Switching Technology, Beijing University of Posts and Telecommunications, Beijing 100876, China;2. School of Software, Beijing Institute of Technology, Beijing 100081, China)
  • Received:2010-01-09 Revised:1900-01-01 Online:2011-01-25 Published:2011-01-25

摘要: 为了提高竞争环境下基于智能体电子商务双边多议题协商当中agent协商的效率,提出了一种竞争环境下agent的协商模型,并且将改进的遗传算法IGA应用于这种模型当中,来提高模型中agent协商的效率。在相同实验条件下,分别对四种遗传算法各进行了1000次的实验,结果表明同样达到协商满意解的时候,标准遗传算法SGA平均需要360次协商,基于Metropolis准则的遗传算法MGA平均需要230次协商,自适应遗传算法AGA平均需要207次协商,而IGA平均仅需要151次协商。这个结果说明,在求解竞争环境下双边多议题协商问题的时候,改进的遗传算法IGA可以使得协商当中的agent高效达到协商的满意解。

关键词: 电子商务, 双边多议题协商, 改进的遗传算法

Abstract: To make the agents negotiate more efficiently in bilateral multi-issue negotiation in multi-agent based competitive e-commerce, an agent negotiation model in competitive environment is presented, and the Improved Genetic Algorithm (IGA) is applied in the model to enhance the negotiation efficiency. After 1000 times of experiments for the four kinds of agents in the same condition, the Standard Genetic Algorithm (SGA) averagely needs negotiation of 360 runs, Genetic Algorithm based on Metropolis rule (MGA) averagely needs 230 runs, Adaptive Genetic Algorithm AGA averagely needs 207 runs, while the IGA averagely needs only 151 runs. The experiment results show that the IGA can gain the optimal negotiation result more efficiently than other three kinds of genetic algorithms in competitive bilateral multi-issue negotiation.

Key words: e-commerce, bilateral multi-issue negotiation, improved genetic algorithm

中图分类号: