运筹与管理 ›› 2022, Vol. 31 ›› Issue (3): 31-37.DOI: 10.12005/orms.2022.0074

• 理论分析与方法探讨 • 上一篇    下一篇

基于先验信息和一维数据聚类的专家赋权方法

易平涛, 王士烨, 李伟伟, 王露, 董乾坤   

  1. 东北大学 工商管理学院,辽宁 沈阳 110167
  • 收稿日期:2020-04-19 出版日期:2022-03-25 发布日期:2022-04-12
  • 作者简介:易平涛(1981-),男,湖南永州人,教授,博士,研究方向:评价理论与方法、信息挖掘;王士烨(1995-),男,山东东营人,博士研究生,研究方向:综合评价。
  • 基金资助:
    国家自然科学基金资助项目(72171040,72171041)

Expert Weighting Method Based on Prior Information and One-dimensional Data Clustering

YI Ping-tao, WANG Shi-ye, LI Wei-wei, WANG Lu, DONG Qian-kun   

  1. School of Business Administration, Northeastern University, Shenyang 110167, China
  • Received:2020-04-19 Online:2022-03-25 Published:2022-04-12

摘要: 多属性群体评价方法研究中,对专家赋权是一项重要的研究内容。本文结合专家的先验信息和后验信息,提出了一种确定专家权重的方法。首先,利用专家历史评价活动中的序值相关系数,做出本次评价活动的预测值,进而确定专家先验权重;其次,基于群体共识视角,对各专家所给出的指标信息进行一维数据聚类,结合不同分组情况下出现的概率,计算出专家后验权重;最后,将两类权重进行组合确定最终权重,并用一个算例验证了该方法的有效性以及合理性。

关键词: 综合评价, 多属性群体评价, 一维数据聚类, 专家权重

Abstract: In the research of multi-attribute group evaluation method, expert empowerment is an important research content. Combining the prior information and the posterior information of experts, this paper presents a method to determine the weight of experts. First of all, the order value correlation coefficient in the expert history evaluation activity is used to make the predicted value of this evaluation activity, and then the prior weight of experts is determined. Secondly, based on the perspective of group consensus, one-dimensional data clustering is carried out for the indicator information given by each expert, and the expert posterior weight is calculated in combination with the probability of occurrence under different grouping conditions. Finally, the two types of weights are combined to determine the final weight, and an example is given to verify the validity and rationality of the method.

Key words: comprehensive evaluation, multi-attribute group evaluation, one-dimensional data clustering, expert weight

中图分类号: