Entropy weight computing method and multi-target decision method

Abstract

The invention discloses an entropy weight calculation method and device. The method includes: obtaining the evaluation index of the evaluated object and the entropy value of each evaluation index; calculating the entropy weight of each evaluation index according to the following formula: &omega; ej = ( 1 - H ‾ ) &omega; ej 1 + H ‾ &omega; ej 2 H j < 1 0 H j = 1 ω ej 1 = 1 - H j Σ j = 1 , H j ≠ 1 m ( 1 - H j ) , ω ej 2 = 1 / H j Σ j = 1 , H j ≠ 1 m ( 1 / H j ) Among them, ω ej Indicates the entropy weight of the evaluation index j, H j Indicates the entropy value of the evaluation index j, Indicates the mean value of all entropy values ​​that are not 1, m indicates the number of evaluation indicators, j=1,2,...,m. As a result, the entropy value can be reasonably weighted when the entropy value is too large or too small, which solves the existing entropy weight method in the prior art. When the entropy values ​​of all evaluation indicators approach 1, the small difference in entropy value will cause the corresponding entropy The weight changes exponentially, which leads to the problem of unreasonable distribution of entropy weight. In addition, the embodiment of the present invention also provides a multi-objective decision-making method and device based on the above technical solution, which is simple and efficient, and is very suitable for online operation.

Description

technical field [0001] The present invention relates to the technical field of entropy weight method, and more specifically, relates to an entropy weight calculation method and a multi-objective decision-making method. Background technique [0002] Entropy itself is a thermodynamic concept, which was later introduced into information theory by C.E. Shannon, endowing entropy with a broad concept. According to the idea of ​​entropy, the quantity and quality of information that people obtain in decision-making is one of the important factors that determine the accuracy and reliability of decision-making. The entropy weight method is an objective weighting method formed by using the characteristic that entropy can measure the amount of useful information provided by data. In the case of a given set of evaluation objects and the values ​​of various evaluation indicators are determined, from the perspective of information, the weight given to an indicator by using the entropy wei...

AI Technical Summary

Problems solved by technology

[0013] The purpose of the present invention is to provide an entropy weight calculation method and device to solve the problem in the existing entropy weight method that when the entropy values ​​of all evaluation indicators approach 1, the slight difference in entropy value will cause the corresponding entropy weight to become times change, which leads to the problem of unreasonable distribution of entropy weights; in addition, the present invention also provides a multi-objective decision-making method and device

Images