Construction Of Mobile User Behavior Analysis System Based On Data Mining

At present,most analysis systems of mobile user behavior use simple local data to model and analyze the behavior of mobile users,which has the problems of low efficiency,poor reliability,and low accuracy of prediction.In view of the above problems,the dissertation studies deeply the basic concepts and research the current states of domestic and foreign and mobile user behavior analysis system,data mining,neural network and genetic algorithm.Based on the analysis model of mobile user behavior and the research of the existing technologies and methods,this dissertation adopts the process of data mining to research and analyze the mobile user behavior,and proposes an improved BP algorithm and applies it to the analysis system of mobile user behavior,which can improve the accuracy of the system analysis and prediction.BP neural network algorithm has low speed of convergence,low precision and easily falling into the local extremum,the dissertation adopts an optimized BP algorithm based on improved PSO algorithm and GA algorithm.The stability of the algorithm can be improved by improving inertia weight formula to adjust speed updating formula in the PSO algorithm.In the later period of optimization,the dissertation applies the crossover operation and mutation operation of GA algorithm to expand the searching space of the particles,which can improve the diversification of the particles and prevent particles from falling into the local solution previously and improve the accuracy of the BP neural network algorithm finally.The results of simulation show the improved BP algorithm not only solves the problems of low efficiency and poor credibility of the traditional analysis system,but also effectively improves the accuracy of the system prediction.To a certain extent,it can provide a theoretical basis for the quality optimization of mobile communication networks and precision marketing,and help network operators improve their network service levels.