主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2024, Vol. 32 ›› Issue (10): 234-243.doi: 10.16381/j.cnki.issn1003-207x.2021.2002cstr: 32146.14.j.cnki.issn1003-207x.2021.2002

• • 上一篇    下一篇

新零售生态网络安全态势评级混合型变权方法及稳定性研究

余高锋1,李登峰2()   

  1. 1.福建商学院工商管理学院, 福建 福州 350016
    2.电子科技大学经济与管理学院, 四川 成都 611731
  • 收稿日期:2021-09-29 修回日期:2022-11-29 出版日期:2024-10-25 发布日期:2024-11-09
  • 通讯作者: 李登峰 E-mail:lidengfeng@uestc.edu.cn
  • 基金资助:
    教育部人文社会科学青年项目(20YJC630196);国家自然科学基金项目(72001126);福建省自然科学基金项目(2020J01925)

The Stability and Variable Weight Decision Method for Rating New Retail Ecology Network Security Situation

Gaofeng Yu1,Dengfeng Li2()   

  1. 1.School of Business Administration, Fujian Business University, Fuzhou 350016, China
    2.School of Management and Economics, University of Electronic Science and Technology of China, Chengdu 611731, China
  • Received:2021-09-29 Revised:2022-11-29 Online:2024-10-25 Published:2024-11-09
  • Contact: Dengfeng Li E-mail:lidengfeng@uestc.edu.cn

摘要:

新零售生态网络安全态势评级是常见问题,但现有网络安全态势评级方法主要针对一般网络安全,对新零售生态网络安全尚未有研究,尤其很少关注指标间信息非补偿性和评级方法稳定性。为此,本文研究了新零售生态网络安全态势评级混合型变权方法及稳定性。在描述了新零售生态网络安全态势评级问题基础上,构造了阈值为区间型、上界型、下界型和语言型等指标特征值级别隶属函数,并且给出四种情形的隶属函数表达式;提出了基于交叉熵和熵理论的指标权重确定方法;据此提出混合型变权的模糊综合评价模型,进一步运用二元语义方法确定等级和区分度,利用数理方法研究新零售生态网络安全态势评级混合型变权综合模型的稳定性,并给出新零售生态网络安全态势评级混合型变权综合计算结果(等级)保持不变时的指标权重变化范围解析解。该方法克服最大隶属度评级方法不足,同时解决现有评级方法的指标间信息补偿性问题。

关键词: 评级, 稳定性, 变权方法, 网络安全

Abstract:

The rating of new retail ecology network security is a key problem in nowadays cybersecurity management. However, the existingnetwork security rating investigated on general network security. New retail ecology network security is not studied, especially paying little attention to issues such as the non-compensation of information among factors and the stability of rating methods, the stability for variable weight decision methods for rating network security situation. Based on the description of the fuzzy rating problem, the membership functions of interval type, upper bound type, and lower bound type are constructed, and the membership functions of four cases are given. A method of determining index weights based on cross-entropy and entropy theory is proposed. A compromised variable weight analysis method is systematically studied. A compromised variable weight function construction method based on utility function is proposed. Furthermore, the absolute risk aversion coefficient is used to analyze the effect of variable weight decision-making, and the relevant properties are proved. Accordingly, a fuzzy comprehensive evaluation model of compromised variable weight is proposed, and the binary semantic method is further used to identify grades reasonably and distinguish the ranking of the same grade. The stability analysis of cybersecurity situation grade assessment mixed variable weight synthesis model is studied through using mathematical theory, and the analytical solutions of index weight ranges are given when the grades’ results of the cybersecurity situation grade assessment mixed variable weight method are stable. This method can overcome the shortcomings of the maximum membership rating method and compensation index rating problems.

Key words: grade assessment, stability, variable weight method, network security

中图分类号: