A Study on Correlation Analysis and Application of Communication Network Service

Abstract

Research on mobile network user behavior is of vital importance for mobile communications services. However, in most cases, large amounts of data are collected in a dispersed way and irregularly, impeding operators from offering reasonable business recommendations for users, hence business managers can hardly satisfy customers’ needs. Therefore, we conduct a correlation analysis of mobile data, including a correlation analysis between a user’s mobile network business and district, between business and time, and between business and business. In the process of analyzing, based on the Hadoop distributed computing platform, we apply the FP-Growth correlation algorithm to analyze real communication data from operators, obtaining the business type transition law of the network service types of users. Then, by applying the K-means algorithm, we cluster the data of base stations in the city, classify and count the network traffic flow used by users in each area and acquire their preference of network access. Furthermore, through the visualization of base station data of network businesses, mobile data could be better applied and analyzed. The study is useful for achieving the optimization of resource allocation and communication network service provided by operators, contributing to the better guidance of urban planning.