Difference Measure Method of Risk Probability Distribution Based on Moment Generating Function and Fuzzy Data Stream Clustering
Difference Measure Method of Risk Probability Distribution Based on Moment Generating Function and Fuzzy Data Stream Clustering
复制标题
基于矩生成函数和模糊数据流聚类的风险概率分布差分测度方法
DOI：
--
复制
发表时间：
2019
影响因子：
1.1
通讯作者：
Cai Bo-Yu
中科院分区：
文献类型：
--
作者：
Xue Hong;Yuan Yi;Jiang Chi;Cai Bo-Yu
风险概率分布差异测度方法的研究在零售供应链非常规突发事件预警决策管理中起着关键作用。而常用的差异性度量指标是根据风险概率分布的特定密度函数或分布规律建立的。在Knight不确定环境下，只能得到风险概率分布的矩.针对零售供应链非常规突发事件，提出了基于矩母函数和模糊数据流聚类的风险概率分布差矩测度方法。对风险评估指标进行大数据统计分析，得到非常规突发事件风险概率分布的矩。利用无穷维矩真实的向量空间中的距离函数度量非常规突发事件风险矩母函数的差异，进而利用矩度量非常规突发事件风险概率的真实的分布与参考分布的差异。本文的主要贡献是在Knight不确定大数据环境下，基于云模型方法、矩母函数理论、函数函数和大数据模糊统计技术，提出了一种新的非常规突发事件风险概率分布差异度量方法，克服了现有概率分布差异度量方法的不足。
The research of the difference measure method for risk probability distribution plays a key role in the early warning decision-making management of retail supply chain unconventional emergency. However, the common difference measure indices are established by the specific density function or distribution law of the risk probability distribution. In Knight uncertain environment, only the moments of the risk probability distribution can be obtained. This study proposes the difference moment measure method of risk probability distribution based on moment generating function and fuzzy data stream clustering for the retail supply chain unconventional emergency. The big data statistical analysis is performed on the risk assessment indices to obtain the moments of the risk probability distribution for unconventional emergency. The difference of moment generating functions for unconventional emergency risk is measured by the distance function in the real vector space of infinite dimensional moments and then the difference between the real distribution and the reference distribution of the risk probability for unconventional emergency is further measured by the moments. The main contribution of this study is that we propose a new difference measure method of risk probability distribution for unconventional emergency based on cloud model method, moment generating function theory, functional function and big data fuzzy statistics technology in Knight uncertain and big date environments, which can overcome the drawbacks of the existing difference measure methods for probability distributions.
登录
查看更多内容
用于降低客户流失风险的大数据聚类算法
影响因子：
12.3
作者：
Wenjie Bi;Meili Cai;Mengqi Liu;Guo Li
通讯作者：
Guo Li
影响因子：
1.6
作者：
I. Alberink;Annette Sprong;A. Bolck;P. Vergeer
通讯作者：
I. Alberink;Annette Sprong;A. Bolck;P. Vergeer
影响因子：
5.5
作者：
R. Yager
通讯作者：
R. Yager
影响因子：
1.5
作者：
Soumita Ghosh;D. Ghosh;D. Mohanta
通讯作者：
Soumita Ghosh;D. Ghosh;D. Mohanta
DOI：
10.1016/j.amc.2016.05.026
发表时间：
2016-11
期刊：
Appl. Math. Comput.
影响因子：
--
作者：
D. Villela
通讯作者：
D. Villela