Application Research Of Fuzzy Clustering Based On Genetic Algorithm In User Behavior Analysis

Telecom user behavior analysis refers to the use of data mining and other related technologies to conduct a scientific analysis of the behavior of telecommunications users in order to obtain some rules of user consumption,and to use these laws and telecommunications Enterprise scientific decision-making,targeted marketing and cross-selling program design combined.It is to improve customer demand for customers,and better serve customers.Fuzzy C-means(FCM)Clustering analysis is an important data processing technology in data mining and is widely studied and applied in various fields.FCM algorithm is easy to understand,description concise,practical,fast convergence and automatic classification and so on.However,in order to solve the problem of FCM algorithm,the FCM clustering algorithm is applied to the FCM clustering algorithm because the genetic algorithm has the ability of global optimization,and the genetic algorithm is easy to use other algorithms combined.Therefore,combining genetic algorithm with FCM clustering analysis,relying on FCM’s strong convergence speed and excellent global optimization ability of genetic algorithm,it can be well applied to telecom user behavior analysis.In this thesis,based on the deep study of FCM clustering algorithm and genetic algorithm,the traditional genetic algorithm and FCM clustering algorithm are optimized to a certain extent.In the optimization of genetic algorithm,the fitness function of genetic algorithm,the selection of genetic operator(selection,crossover and mutation)and the selection of genetic parameters are optimized.In FCM clustering algorithm optimization,the related parameters(fuzzy factor)were optimized.By combining the optimized genetic algorithm with FCM clustering analysis and applying it to the behavior analysis of telecommunication users,the telecom user behavior involved in this paper is mainly the behavior of telecom users(call behavior,SMS behavior and traffic behavior).Depth analysis of these telecommunications user behavior,and then provide a reasonable reference for the proposed package,and for scientific decision-making,targeted marketing and cross-selling program design basis.And the feasibility of the algorithm is analyzed by experimental analysis.