基于乘积偏好关系的专家模糊核聚类赋权方法
发布时间：2018-07-24 07:48
【摘要】：多属性、多目标性决策中,针对专家给出各方案偏好关系下的决策问题,提出一种基于乘积偏好关系的专家模糊核聚类赋权方法。该方法运用模糊核聚类的思想实现对决策专家的聚类,并通过放宽归一化约束条件,克服了传统模糊核聚类算法中离群点对聚类结果的影响。同时,在专家类内赋权过程中,运用CI-IOWG算子集结同类专家的意见,依据不同专家对于形成类别一致性意见的贡献程度来确定专家权重;克服了传统基于熵权或判断矩阵一致性的赋权方法的局限性。算例表明,该方法可行、有效。
[Abstract]:In multi-attribute and multi-objective decision making, an expert fuzzy kernel clustering method based on product preference relation is proposed to solve the problem of expert preference relationship. This method uses the idea of fuzzy kernel clustering to realize the clustering of decision experts, and by relaxing the normalized constraint conditions, it overcomes the influence of outliers on the clustering results in the traditional fuzzy kernel clustering algorithm. At the same time, in the process of experts' intra-class weighting, the expert weight is determined according to the contribution of different experts to the formation of category consistency by using CI-IOWG operator to gather the opinions of the same kind of experts. It overcomes the limitation of traditional weighting method based on entropy weight or consistency of judgment matrix. An example shows that this method is feasible and effective.
【作者单位】： 空军工程大学装备管理与安全工程学院;
【基金】：“十二五”国防预研基金资助项目(51327020104)
【分类号】：O225
本文编号：2140705
[Abstract]:In multi-attribute and multi-objective decision making, an expert fuzzy kernel clustering method based on product preference relation is proposed to solve the problem of expert preference relationship. This method uses the idea of fuzzy kernel clustering to realize the clustering of decision experts, and by relaxing the normalized constraint conditions, it overcomes the influence of outliers on the clustering results in the traditional fuzzy kernel clustering algorithm. At the same time, in the process of experts' intra-class weighting, the expert weight is determined according to the contribution of different experts to the formation of category consistency by using CI-IOWG operator to gather the opinions of the same kind of experts. It overcomes the limitation of traditional weighting method based on entropy weight or consistency of judgment matrix. An example shows that this method is feasible and effective.
【作者单位】： 空军工程大学装备管理与安全工程学院;
【基金】：“十二五”国防预研基金资助项目(51327020104)
【分类号】：O225
【相似文献】
相关期刊论文 前10条
1 张国权;李文立;王明征;;基于离差函数和联合熵的组合赋权方法[J];管理学报;2008年03期
2 李明奇;刘玉娟;赵美玲;;一种基于判断矩阵的专家赋权方法[J];科技信息;2010年21期
3 席雪红;;基于熵—Shapely的样本差异赋权方法[J];统计与决策;2012年20期
4 许叶军;达庆利;;不确定有序加权几何平均算子的赋权方法(英文)[J];Journal of Southeast University(English Edition);2008年01期
5 申锦标;吕跃进;;一种基于向量贴近度的组合赋权方法[J];重庆工学院学报(自然科学版);2009年02期
6 张玉;魏华波;;基于CRITIC的多属性决策组合赋权方法[J];统计与决策;2012年16期
7 陈琼华;综合评价中的赋权方法[J];统计与决策;2004年04期
8 杨超,李志强;基于主成分—神经网络相集成的多策略评价赋权方法[J];当代财经;2005年07期
9 朱建军;吴伟丽;刘思峰;;一种基于模糊线性规划的主客观组合赋权方法[J];运筹与管理;2006年03期
10 ;[J];;年期
相关会议论文 前1条
1 郑凯;;对统计分析中常用赋权方法及权值效用的分析[A];第七届全国体育科学大会论文摘要汇编（二）[C];2004年
，本文编号：2140705
本文链接：https://www.wllwen.com/kejilunwen/yysx/2140705.html
