Font Size: a A A

Design And Implementation Of Telecom Customer Churn Prediction System

Posted on:2023-04-24Degree:MasterType:Thesis
Country:ChinaCandidate:S W WuFull Text:PDF
GTID:2568307061963919Subject:Software engineering
Abstract/Summary:PDF Full Text Request
In recent years,the state’s supervision of the telecommunications industry has become increasingly strict.At the same time,the telecommunications industry is facing increasingly fierce market competition.In this context,it is particularly important to predict customer churn in order to formulate targeted customer marketing strategies and increase customer stickiness.Starting from the core needs of telecom operators,this paper designs and implements a telecom customer churn prediction system,collects and processes relevant customer related data,applies customer value clustering model and customer churn prediction model to predict customer churn,so that enterprises can retain high-value customers and reduce customer churn.The main work of this paper is as follows:(1)Data acquisition and preprocessing.Kettle is used to extract,transform and monitor data,and then perform data preprocessing including data cleaning,feature construction,data normalization,etc.,and propose a missing value filling method based on KNN,and the feature selection of customer churn data set is carried out,so as to improve the quality of training data.(2)Construction of customer churn prediction model.Using entropy weight method and correlation analysis method to determine the index weight of model parameters,a telecom customer segmentation method based on weighted k-means algorithm is proposed;Using logistic regression,random forest,xgboost and lightgbm algorithms to build customer churn prediction models respectively,and optimize the model parameters,then select the model with high AUC for Voting fusion methods(including soft voting and hard voting),and apply the clustering results to churn prediction to improve the prediction accuracy.(3)Design and implement a telecommunication customer churn prediction system.System development is based on Spring Boot and Vue frameworks.The system mainly includes system basic function module,data acquisition module,data modeling module and data display module,and the functional and non functional tests of the system are carried out to verify the effectiveness and reliability of the system.
Keywords/Search Tags:KNN, Customer segmentation, Weighted K-Means, Churn prediction, Voting fusion
PDF Full Text Request
Related items