计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (17): 19-21.DOI: 10.3778/j.issn.1002-8331.2009.17.006

• 博士论坛 • 上一篇    下一篇

声望模型中一种动态选择遗忘因子的方法

贡佳炜1,2,单明辉1,2,陈 君2,邓浩江2,王劲林2   

  1. 1.中国科学院 研究生院,北京 100049
    2.中国科学院 声学研究所 国家网络新媒体工程技术研究中心,北京 100190
  • 收稿日期:2009-01-07 修回日期:2009-03-11 出版日期:2009-06-11 发布日期:2009-06-11
  • 通讯作者: 贡佳炜

Method of dynamically selecting forgetting factor in reputation model

GONG Jia-wei1,2,SHAN Ming-hui1,2,CHEN Jun2,DENG Hao-jiang2,WANG Jin-lin2   

  1. 1.Graduate University of Chinese Academy of Sciences,Beijing 100049,China
    2.National Network New Media Engineering Research Center,Institute of Acoustics,CAS,Beijing 100190,China
  • Received:2009-01-07 Revised:2009-03-11 Online:2009-06-11 Published:2009-06-11
  • Contact: GONG Jia-wei

摘要: 针对目前声望模型中单一的遗忘因子无法准确地跟踪动态变化的声望值的问题,提出了一种以降低总误差为目标的动态选择遗忘因子的方法。该方法首先分析了不同的遗忘因子对总误差的影响,然后以声望值的变化程度为依据,在变化较为剧烈时选择较大的遗忘因子以快速体现变化,在变化较小时选择较小的遗忘因子以减小随机误差。仿真结果表明:该方法是行之有效的。

关键词: 声望模型, 遗忘因子, 动态选择

Abstract: In order to address the problem that most current reputation models cannot track the change of reputation accurately,this paper proposes a method of dynamically selecting forgetting factor to minimize the total error.After analyzing the influence of different forgetting factor on total error,it is based on the change degree of reputation to select forgetting factor.When reputation changes sharply,it selects the bigger forgetting factor to track the change quickly.Otherwise,it selects the smaller forgetting factor to reduce the random error.The simulation results demonstrate that this method is effective.

Key words: reputation model, forgetting factor, dynamic selection