Anatomy Of Adverse Judgment Problem On Group Decision Making Based On Fuzzy Soft Set

For dealing with uncertain decision-making problems,traditional methods such as fuzzy set and rough set can be used.However,these methods have their own shortcomings.For example,fuzzy set maybe restricted by its membership function,and the membership function will affect the decision-making results.In recent years,fuzzy soft set has become a hot spot,which has its own specific advantages in dealing with uncertain information group decision making problems.Therefore,the analysis of inverse decision making based on fuzzy soft set has become a focus problem,and how to select the non-cooperative expert groups in large-scale group decision making has become a difficult problem.This paper mainly studies the inverse decision making based on fuzzy soft set in group decision making.By combining the clustering analysis algorithm,the inverse decision making method based on fuzzy clustering algorithm and K-means algorithm are given,and comparing with the traditional algorithm.By calculating the algorithm complexity,it can be concluded that using clustering algorithm to deal with the inverse judgement problem of fuzzy soft sets can select the non-cooperative expert group more quickly,and reduce the time and the space complexity of the calculation.Firstly,the research background and current situation of group decision-making,inverse decision-making,fuzzy soft set and fuzzy clustering are introduced,and the definitions and operation rules of fuzzy soft set,fuzzy soft matrix and similarity measure are summarized.Secondly,a fuzzy soft judgment matrix is given to calculate the similarity measure of fuzzy soft matrix.Then,two algorithms are proposed,one is based on fuzzy clustering,the other one is based on K-means algorithm,using these two methods to calculate the inverse judgement problem of fuzzy soft sets.Then,to classify the problem,and selects an expert from each category,calculates the consistency ratio of the experts,and obtains the judgement level of the experts' classes.Then,an example is given to illustrate the feasibility of the two methods.After summarizing the full text,the advantages of using clustering algorithm for large-scale decision-making are illustrated,the shortcomings are summarized and the prospects are put forward.