Research And Application Of Intuitionistic Fuzzy Cluster Analysis Based On Entropy Weight Similarity

With the continuous development of human society,the classification of nature has become more complicated.Some subjects do not have specific attributes,the nature of things tend to be neutral,the classification of such things must be accompanied by ambiguity.Fuzzy mathematics has provided a good theoretical basis for this classification problem.Combining fuzzy set theory with cluster analysis has promoted the development of classification problem,and more and more experts and scholars have been engaged in this kind of research.But the fuzzy sets of the research object fuzzy degree of characterization is not comprehensive enough.In order to fully exploit the effective information of the data and make up the deficiency of the fuzzy set,Atanassov extended the fuzzy set theory to the intuitionistic fuzzy set theory in 1986,which increased a new attribute parameter and more comprehensively depict the uncertainty of the objective world.Fuzzy clustering analysis also extends to the intuitionistic fuzzy clustering analysis.In this paper,the exploratory research is carried out from the following aspects.The information entropy of fuzzy sets can describe the fuzzy degree of fuzzy sets.Based on the concept and definition of intuitionistic fuzzy entropy,this paper explains the intuitionistic fuzzy entropy from the geometric point of view,and proposes a new intuitionistic fuzzy entropy formula.The main idea is the ratio of the distance between the distance of any intuitionistic fuzzy point to the minimum point of the information entropy and the sum of the distance of the maximum and minimum points of the information entropy as the size of the intuitionistic fuzzy entropy of the intuitionistic fuzzy point.And formulas were normalized so that the formula specifications and reasonable.Secondly,this paper constructs a new intuitionistic fuzzy entropy formula based on ambiguity and hesitation.The formula is simple and easy to operate,and the characterization of the object is better described.This paper also proposes a new method of constructing intuitionistic fuzzy numbers as intuitionistic fuzzy similarity.The main idea is to take the minimum value of the distance between the membership degree and the non-membership degree of the fuzzy set as the non-membership degree of similarity.And then we take the difference of one misusing the maximum value of the distance between the membership degree and the non-membership degree as the membership degree.The calculation formulatakes into account the different degree of contribution of the different indicators to the results,increasing the weight coefficient of the index,making the calculation results more in line with the practical significance.The form of the calculation formula is simple and well reflects the close degree of the research object,which lays the foundation for the subsequent intuitionistic fuzzy clustering analysis.Finally,this paper to 20 air targets for classification as a example,and examines the seven attribute indicators of the research object.The author uses the intuitionistic fuzzy entropy formula proposed in Chapter 3 to determine the attribute weight of each index,and then use the intuitionistic fuzzy similarity formula proposed in Chapter 4 to calculate the weighted similarity of each two air targets.The clustering algorithm of maximum tree and equivalence relation is used to analyze and obtain the same result as expert prediction,and the reliability of the proposed algorithm is explained.