GAN data reconstruction based prediction method of telecom subscriber loss

Kehong A; Xiaodong HU

doi:10.11959/j.issn.1000-0801.2023038

您当前的位置：

首页 >

文章列表页 >

GAN data reconstruction based prediction method of telecom subscriber loss

Research and Development | 更新时间：2024-06-05

- GAN data reconstruction based prediction method of telecom subscriber loss
- Telecommunications Science Vol. 39, Issue 3, Pages: 135-142(2023)
- 作者机构：
- 作者简介：
- 基金信息：
- DOI：10.11959/j.issn.1000-0801.2023038
  CLC：
- Published Online：2023-03，
  
  Published：20 March 2023
- 稿件说明：
移动端阅览
Kehong A, Xiaodong HU. GAN data reconstruction based prediction method of telecom subscriber loss[J]. Telecommunications science, 2023, 39(3): 135-142.
DOI：

Kehong A, Xiaodong HU. GAN data reconstruction based prediction method of telecom subscriber loss[J]. Telecommunications science, 2023, 39(3): 135-142. DOI： 10.11959/j.issn.1000-0801.2023038.

摘要

用户是运营商利益的核心。随着携号转网政策的出台，运营商之间的竞争越发激烈。为了提前精准有效地预测用户流失倾向，提出了一种基于生成对抗网络（generative adversarial network，GAN）数据重构的电信用户流失预测方法。首先，利用有效的数据预处理方法电信用户流失数据中的脏数据；其次，利用GAN重构电信用户流失数据，解决电信用户流失数据不平衡问题；最后，利用极度梯度提升树（extreme gradient boosting，XGBoost）算法分别训练基于 GAN 重构的电信用户流失预测模型和基于合成少数类过采样技术（synthetic minority oversampling technique，SMOTE）采样的电信用户流失预测模型，对比两种模型的预测精度。实验结果表明，GAN 重构后的电信用户流失预测模型预测精度比未重构的预测模型的准确率提升了6.75%，查准率提升了25.91%，召回率提升了30.91%，F1值提升了28.73%。该方法能够有效提升电信用户流失预测的准确度。

Abstract

Users are the core of operators’ interests.With the introduction of the policy of transferring network with a number

the competition between operators becomes more and more fierce.In order to accurately predict subscriber loss tendency in advance

a prediction method of subscriber loss based on generative adversarial network data reconstruction was proposed.Firstly

the dirty data in the telecom subscriber loss data was used by effective data preprocessing method.Secondly

the GAN was used to reconstruct the telecom subscriber loss data to solve the problem of the imbalance of the telecom subscriber loss data.Finally

extreme gradient boosting algorithm was used to train the telecom subscriber loss prediction model based on GAN reconstruction and the SMOTE sampling model based on synthetic minority oversampling technique sampling method respectively

and compare the prediction accuracy of the two models.The experimental results show that the prediction accuracy of the GAN reconstructed telecom subscriber loss prediction model is increased by 6.75%

the accuracy rate is increased by 25.91%

the recall rate is increased by 30.91%

and the F1-score is increased by 28.73% compared with the unreconstructed prediction model.This method can effectively improve the accuracy of telecom subscriber loss prediction.

关键词

Keywords

references

叶艳艳 . 基于机器学习的电信用户流失预警模型预测与分析 [D ] . 济南:山东大学 , 2021 .

YE Y Y . Prediction and analysis of telecom customer churn warning model based on machine learning [D ] . Jinan:Shandong University , 2021 .

钟保权 . 基于图学习的电信用户流失预测算法 [D ] . 杭州:浙江大学 , 2021 .

ZHONG B Q . Graph learning algorithm for telecom user churn prediction [D ] . Hangzhou:Zhejiang University , 2021 .

乔健 , 诸佳慧 , 严康桓 . 基于随机森林 CART 特征选择改进算法的电信客户流失预测模型 [J ] . 电信工程技术与标准化 , 2022 , 35 ( 3 ): 78 - 82 .

QIAO J , ZHU J H , YAN K H . Telecom customer churn prediction model based on improved random forest cart feature selection algorithm [J ] . Telecom Engineering Technics and Standardization , 2022 , 35 ( 3 ): 78 - 82 .

ZHANG T Y , MORO S , RAMOS R F . A data-driven approach to improve customer churn prediction based on telecom customer segmentation [J ] . Future Internet , 2022 , 14 ( 3 ): 94 .

JAIN H , KHUNTETA A , SRIVASTAVA S . Telecom churn prediction and used techniques,datasets and performance measures:a review [J ] . Telecommunication Systems , 2021 , 76 ( 4 ): 613 - 630 .

郑声晟 , 殷海兵 , 黄晓峰 , 等 . 基于GAN的无监督域自适应行人重识别 [J ] . 电信科学 , 2021 , 37 ( 2 ): 99 - 106 .

ZHENG S S , YIN H B , HUANG X F , et al . GAN-based unsupervised domain adaptive person re-identification [J ] . Telecommunication Science , 2021 , 37 ( 2 ): 99 - 106 .

ERIA K , MARIKANNAN B P . Significance-based feature extraction for customer churn prediction data in the telecom sector [J ] . Journal of Computational and Theoretical Nanoscience , 2019 , 16 ( 8 ): 3428 - 3431 .

LIU H Z , ZHANG X , SHEN X W , et al . A fair and efficient hybrid federated learning framework based on XGBoost for distributed power prediction [J ] . arXiv preprint , 2022 ,arXiv:2201.02783.

王愈轩 , 梁沁雯 , 章思远 , 等 . 基于 LSTM-XGBoost 组合的超短期风电功率预测方法 [J ] . 科学技术与工程 , 2022 , 22 ( 14 ): 5629 - 5635 .

WANG Y X , LIANG Q W , ZHANG S Y , et al . An ultra-short-term wind power prediction method based on LSTM-XGBoost combination [J ] . Science Technology and Engineering , 2022 , 22 ( 14 ): 5629 - 5635 .

LIN M Y , ZHU X F , HUA T , et al . Detection of ionospheric scintillation based on XGBoost model improved by SMOTE-ENN technique [J ] . Remote Sensing , 2021 , 13 ( 13 ): 2577 .

HAN H , WANG W Y , MAO B H . Borderline-SMOTE:a new over-sampling method in imbalanced data sets learning [M ] // Lecture Notes in Computer Science . Heidelberg : Springer , 2005 : 878 - 887 .

张晨路 . 基于G-SMOTE 和Biased-SVM 的内部威胁用户检测 [J ] . 中北大学学报(自然科学版) , 2022 , 43 ( 2 ): 147 - 152 .

ZHANG C L . User detection of insider threat detection based on G-SMOTE and biased-SVM [J ] . Journal of North University of China (Natural Science Edition) , 2022 , 43 ( 2 ): 147 - 152 .

石洪波 , 陈雨文 , 陈鑫 . SMOTE过采样及其改进算法研究综述 [J ] . 智能系统学报 , 2019 , 14 ( 6 ): 1073 - 1083 .

SHI H B , CHEN Y W , CHEN X . Summary of research on SMOTE oversampling and its improved algorithms [J ] . CAAI Transactions on Intelligent Systems , 2019 , 14 ( 6 ): 1073 - 1083 .

Views

417

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Unsupervised intrusion detection model based on temporal convolutional network

Ultra-wideband digital channel modeling based on generative adversarial network

Image synthesis method based on multiple text description

Modulation recognition driven by signal enhancement

Adversarial example generation method for SAR images based on mask extraction

Related Author

FENG Guanghui

DING Jiawei

LIAO Jinju

XU Yueqian

ZHUGE Bin

WANG Zhengxian

WANG Ying

CAI Xiaodan

Related Institution

School of Information Engineering, Zhengzhou University of Industrial Technology

School of Computer Science and Cyber Engineering, Guangzhou University

College of Information and Electronic Engineering, Zhejiang Gongshang University

UTStarcom

YingXian School of Philanthropy, Zhejiang Gongshang University

⁰