基于多重属性和证据理论的指标性能评估
首发时间:2020-12-29
摘要:(目的)为保证评价结果的科学性与可靠性,科学进行指标性能评估是指标筛选、赋权及综合评价的基础。(方法)针对现有研究多依据信息量或相关性评估指标性能的单一性,提出一种多重属性信息融合的指标性能评估方法,总结考察评价指标重要性的四个度量尺度:独立度、均衡度、趋优度及灵敏度,结合证据理论的信息融合思想,评估指标综合性能及其重要性。(结果)基于一个算例,计算并对比多重属性信息融合与现有"信息量-相关性"评估方法、TOPSIS信息融合方法的结果。(结论)通过方法对比,验证多重属性信息融合评估的有效性以及证据理论应用的可行性。
For information in English, please click here
Comprehensive evaluation of indicators based on multiple attributes and evidence theory
Abstract:(Object) In order to ensure the scientificity and reliability of the evaluation results, scientific evaluation of indicator performance is the basis for indicator selection, weighting and comprehensive evaluation. (Method) For the one-sidedness of index importance evaluation based on information content or relevance in existing research,this article proposes an index performance evaluation method for multi-attribute information fusion. The method summarizes four measurement aspects to examine the importance of evaluation indicators: Independence, equilibrium, superiority and sensitivity. Combining the information fusion ideas of evidence theory, this article synthesizes the index performance and evaluates its importance. (Result) Based on an example, calculate and compare the results of multiple attribute information fusion with the existing "information-correlation" evaluation method and TOPSIS information fusion method. (Conclusion) The above method verifies the effectiveness of multi-attribute information fusion assessment and the feasibility of the application of evidence theory.
Keywords: Independence equilibrium superiority sensitivity evidence theory
引用
No.****
同行评议
共计0人参与
勘误表
基于多重属性和证据理论的指标性能评估
评论
全部评论0/1000