Design And Application Of FCM Clustering Algorithm For Complex Large Groups

Group decision-making is an important theoretical method for solving various decisionmaking problems.The results of group decision-making can gather opinions from all kinds of sides and make the decision-making results more scientific and reasonable.The size of decision-making groups,however,continues to grow with the continuous development of the Internet economy,and the main body of group decision-making has changed from a few professional experts to a larger group of participants.The heterogeneity of different individuals and the measurable attributes are increasingly diversified.The group decision-making is continuously developed in the direction of decision-making of complex multi-attribute large group.The traditional decision-making theory can not solve the current decision-making problem completely effectively.In order to solve this problem,this paper improves the FCM clustering algorithm so that it can transform complex large group decision-making problems into small-scale group decision-making problems,and applies research on the improved algorithm.The main research ideas of this paper are as follows: The first chapter reviews the background of complex large groups and the current research status,introduces the main research contents and the overall structure of the paper.The second chapter relates to the concept and method theory of complex large group decision-making.The third chapter determines the preprocessing method of complex large group FCM clustering,and gives the streamlined process of streamlining and re-synthesis of data.At the same time,combined with the characteristics of complex large group data,the selection rules of initial clustering center point are determined.Chapter four introduces the evaluation method of distance fusion between classes in FCM clustering process,and proposes an improved FCM clustering algorithm based on Gaussian kernel function.Chapter five combines the theoretical methods of the above chapters to the research of complex large groups of P2 P online lending platforms in FCM clustering algorithm;The sixth chapter summarizes the main work of this paper and gives the prospect of the next stepsThe main contributions of the paper are:1、Aiming at the insensitivity problem caused by the large amount of data in the clustering process of complex large groups,proposing the data preprocessing method which improves the accuracy of the complex large group clustering algorithm and selection rules for initial clustering center points.2、Improving the distance fusion method of the clustering algorithm model and introducing the Gaussian kernel function,which improves the ability of fusion of the distance between the algorithm classes and the distance within the class.3、Clustering analysis of complex large group of borrowers with complex characteristics of P2 P online loan character platform to verify the application analysis ability of the improved algorithm proposed in this paper.