JournalVolume 23, №2Maximum entropy-based method for extracting the underlying probability distributions of z-number (pp.201-218) D. Qiao1,4, X.K. Wang2, J.Q. Wang1,∗, L. Li3 https://doi.org/10.30546/1683-6154.23.2.2024.2011School of Business, Central South University, Changsha 410083, China 2College of Management, Shenzhen University, Shenzhen, 518060, China, e-mail: wangxk93@szu.edu.cnn 3School of Business, Hunan University, Changsha 410082, China 4School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, Shenzhen 518172, China ∗Corresponding author e-mail: jqwang@csu.edu.cn
Abstract. Information reliability has a crucial impact on the accuracy of decision-making. Zadeh’s Z-number is a normalized structure that takes into account the ambiguity and reliability of decision event. However, certain issues such as the extraction of underlying probability distributions (UPDs) of Z-number is somehow intractable due to the complicated structure of Z-number. This study focuses on the extraction of the UPDs of discrete and continuous Z-numbers. The key steps are outlined here. Firstly, a nonlinear programming model based on maximum entropy for extracting the UPDs of discrete Z-number is constructed. Secondly, by combining the Lagrange multiplier method and the maximum entropy method, a method for extracting the UPDs of continuous Z-number is developed. Thirdly, this research introduces the cumulative probability distributions of the UPDs of discrete and continuous Z-numbers. Finally, some numerical examples are used to illustrate the applicability and effectiveness of the proposed methods. Keywords: Z-Number, Underlying Probability Distribution, Maximum Entropy, Lagrange Multiplier Method. AMS Subject Classification: 90-xx. |
|