节点文献

基于距离测度的D-S证据融合决策方法

A Decision-Making Method on D-S Evidence Fusion Information Based on Distance Measure

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 林志贵徐立中严锡君黄凤辰刘英平

【Author】 Lin Zhigui~ 1,2 , Xu Lizhong~ 2 , Yan Xijun~ 2 , Huang Fengchen~ 2 , and Liu Yingping~ 3 ~ 1 (School of Information & Communication Engineering, Tianjin Polytechnic University, Tianjin 300160) ~ 2 (College of Computer and Information Engineering, Hehai University, Nanjing 210098) ~ 3 (College of Water Conservancy and Hydropower Engineering, Hehai University, Nanjing 210098)

【机构】 天津工业大学信息与通信工程学院河海大学计算机及信息工程学院河海大学水利水电工程学院 天津300160河海大学计算机及信息工程学院南京210098南京210098

【摘要】 针对多源信息融合结果并依据D-S证据结构进行决策的问题,提出一种基于距离测度的D-S证据决策方法·该方法结合决策基元和非决策基元的属性进行决策,将依据D-S证据的决策问题分解成两个层面:属性层面、证据层面·属性层面上,给出候选决策从证据焦元获得支持度的方法;证据层面上,基于辨识框架幂集的元素,构造一个证据焦元向量空间,引入候选决策的理想状态向量,定义距离测度,构建决策模型·最后,对多源水质监测信息融合结果进行决策分析,结果表明该方法是合理、有效的,且具有处理冲突或非冲突证据的优点·

【Abstract】 To solve the decision-making problem based on results of multi-source information fusion and D-S evidence structure, a decision-making method based on distance measure is proposed. The method combines attributes of decision-making basic elements with that of decision-making non-basic elements to make a decision about the results, and divides the problem on D-S evidence into two levels, that is, attribute level and evidence level. On the attribute level, a method for obtaining support degrees of candidate decision-making focal elements from evidence focal elements is introduced. On the evidence level, a space based on power set elements of the discernment frame is generated, ideal state vectors of candidate decision-making focal elements are introduced, distance measure is defined, and a decision-making model based on the measure is made. Finally, the fusion results of multi-source information about water quality monitoring data are analyzed based on different methods. The result indicates that the decision-making method based on distance measure is valid and has advantages of processing conflict evidence or non-conflict evidence.

【基金】 国家自然科学基金项目(60374033);江苏省自然科学基金项目(BK2002064)
  • 【文献出处】 计算机研究与发展 ,Journal of Computer Research and Development , 编辑部邮箱 ,2006年01期
  • 【分类号】TP202
  • 【被引频次】22
  • 【下载频次】534
节点文献中: