Research On Three-way Decision Model And Its Applications Under Hesitant Fuzzy Environment

In recent years,the popularity of the Internet and smart devices has led to the emergence of more and more data.With the increasing impact of data on people's lives,data is regarded as the "new oil".How to extract effective knowledge from data is an urgent problem to be studied.As a new extended model of fuzzy sets,hesitant fuzzy set theory is an effective tool to deal with uncertain information in decision problems.Traditional multi-attribute decision making(MADM)method can only choose the optimal strategy and cannot be applied to all complex cases.Usually,in practical applications,the decision problem may be more complex,requiring the decision makers to give three kinds of decisions: acceptance,non-commitment and rejection.Three-way decisions can solve these problems more effectively by using a pair of thresholds to divide the universe into three disjoint regions.Therefore,how to use three-way decision theory to deal with hesitant fuzzy information is a hot research issue.In this thesis,two kinds of hesitant fuzzy three-way decision models are proposed by applying the threedecision theory to the processing of hesitant fuzzy information.The main research contents are as follows:(1)From the perspective of hesitant fuzzy distance,this thesis proposes a hesitant fuzzy three-way decision model.First,since the hesitant fuzzy element is a set of several possible membership degrees,it cannot be directly compared with the thresholds.Therefore,it is converted into the comparison between distance and the threshold in this thesis.Then,in order to calculate the threshold more reasonably,the shadowed set theory is introduced to avoid the subjectivity of threshold acquisition.From the perspective of hesitant fuzzy distance,new decision rules are proposed to divide the universe into three disjoint regions.On this basis,the sequential strategy is used to solve the multi-attribute decision making problem.Finally,the accuracy and effectiveness of the proposed hesitant fuzzy three-way decision model are verified by a medical diagnosis example and a simulation experiment.(2)Combining with Bayesian minimum-risk decision,a new hesitant fuzzy threeway decision model is proposed from the perspective of minimum decision cost.However,because of the special representation of membership of fuzzy sets,we define acceptance degree and deviation degree,which describe the degree of object belonging to the target concept and the degree of object not belonging to the target concept.By means of bayesian decision theory,the degree of acceptance and the degree of deviation are substituted for membership degree and non-membership degree in traditional decision-theoretic rough sets respectively,so as to deduce decision rules.By this method,we get a new calculation formula and a new evaluation index of the threshold value in the hesitant fuzzy environment.In order to make up for the deficiency of the maximum deviation method in the calculation of attribute weight,a method of attribute weight calculation based on the fuzzy entropy of single element in hesitant fuzzy element was proposed to obtain the attribute with stronger partitioning ability.Finally,a three-way hesitant fuzzy decision model based on the minimum cost is proposed,and its effectiveness is verified by experiments.