基于集成学习的广电客户流失预测模型研究
Research on Customer Churn Prediction Model in Broadcasting and Television Industry Based on Ensemble Learning
汪昱霖1
作者信息
- 1. 福州大学 先进制造学院,福建 泉州 362200
- 折叠
摘要
在互联网时代,广电行业面临的竞争日益激烈.针对目前客户流失预测模型精准率不高的问题,提出一种基于集成学习的广电客户流失预测模型.结合客户数据特性,分析不同模型的训练原理差异,通过Stacking集成框架,融合多个具有互补优势的单一预测模型,实现对客户流失的精准预测.在公开数据集上进行仿真验证,并将仿真结果与几种单一模型预测结果进行对比.结果表明,所提模型能够有效提升客户流失的预测精度,有助于广电运营商的客户维系工作.
Abstract
In the Internet era,the broadcasting industry faces increasingly fierce competition.Aiming at the problem of low accuracy rate in the current customer churn prediction model,a customer churn prediction model for broadcasting based on integrated learning is proposed.Combining the characteristics of customer data,analyzing the differences in the training principles of different models,and through the Stacking integration framework,fusing multiple single prediction models with complementary advantages to achieve accurate prediction of customer churn.Simulations are conducted on the public dataset to verify the results,and the simulation results are compared with the prediction results of several single models.The results show that the proposed model can effectively improve the prediction accuracy of customer churn,which is helpful for the customer retention work of broadcasting operators.
关键词
集成学习/客户流失预测/广播电视Key words
ensemble learning/customer churn prediction/radio and television引用本文复制引用
出版年
2024
NETL
NSTL
万方数据