[1]胡永培,张 琛.基于 AP 聚类与随机森林的客户流失预测研究[J].计算机技术与发展,2021,31(02):49-53.[doi:10. 3969 / j. issn. 1673-629X. 2021. 02. 009]
　HU Yong-pei,ZHANG Chen.Research on Prediction Model of Customer Churn Based on AP Clustering and Random Forest[J].,2021,31(02):49-53.[doi:10. 3969 / j. issn. 1673-629X. 2021. 02. 009]
点击复制

基于 AP 聚类与随机森林的客户流失预测研究()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
31
期数:
2021年02期
页码:
49-53
栏目:
大数据分析与挖掘
出版日期:
2021-02-10

文章信息/Info

Title:
Research on Prediction Model of Customer Churn Based on AP Clustering and Random Forest
文章编号:
1673-629X(2021)02-0049-05
作者:
胡永培1; 张 琛2
1. 徽商银行 大数据部,安徽 合肥 230601;?
2. 合肥学院 人工智能与大数据学院,安徽 合肥 230601
Author(s):
HU Yong-pei1; ZHANG Chen2
1. Department of Big Data,Huishang Bank,Hefei 230601,China;?
2. School of Artificial Intelligence and Big Data,Hefei University,Hefei 230601,China
关键词:
客户流失; AP 聚类; CART 决策树; 随机森林; 预测模型
Keywords:
customer churn; AP clustering; CART; random forest; prediction model
分类号:
TP391. 9
DOI:
10. 3969 / j. issn. 1673-629X. 2021. 02. 009
摘要:
利率市场化、大数据迅速发展,银行业均表现出明显的“二八定律”现象,20% 的优质客户占据了银行的大部分资产。那么,如何防止银行客户流失,尤其是优质客户的流失,已经成为银行越来越关注的问题。因此,建立优质客户流失预警模型就显得尤为重要。以某商业银行为例,重新对客户流失进行定义,重点关注银行优质客户的流失预警, 首先使用AP聚类算法进行属性选择,然后使用随机森林方法建立客户流失预警模型,预测零售优质客户未来 3 个月流失的可能性。为了验证该方法的有效性,首先在 UCI 数据集上进行验证,得到了较好的效果,然后使用该方法构建银行业优质客户流失预测模型,实验结果表明该模型的实际预测效果相较于一般的决策树方法,具有更高的准确性。
Abstract:
With the rapid development of interest rate marketization and big data,the banking industry has shown a clear “the 80/20 Rule” phenomenon. 20% of customers occupy most of the bank爷 s assets. Therefore, how to prevent the loss of bank customers, especially the loss of high-quality customers,has become a growing concern of banks. It is particularly important to establish the model of high quality customer loss warning. Taking a commercial bank as an example,we redefine customer churn according to? the actual bank marketing operation and focus on the bank’s high-quality customers. Firstly,AP clustering algorithm is used for attribute selection,and then random forest method is used to establish an early warning model of customer loss,so as to predict? ? the possibility of customer loss of retail high-quality customers in the next three months. In order to verify the effectiveness of the proposed method,it is firstly verified on the UCI data set and a ideal result is obtained. Then,the proposed method is used to structure the prediction model of Bank Customer Churn. The experiment shows that the actual prediction effect of this model is more accurate than that of the general decision tree method.

相似文献/References:

[1]杨杉 何跃.数据仓库和数据挖掘技术在保险公司中的应用[J].计算机技术与发展,2011,(06):157.
　YANG Shan,HE Yue.Application of Data Warehouse and Data Mining to Life Insurance Company[J].,2011,(02):157.
[2]王荣波,王亚杰,黄孝喜,等.基于多算法融合的移动通信客户流失预测模型[J].计算机技术与发展,2018,28(08):152.[doi:10.3969/ j. issn.1673-629X.2018.08.032]
　WANG Rong-bo,WANG Ya-jie,HUANG Xiao-xi,et al.Customer Churn Prediction Model of Mobile Communication Based on Multi-algorithm Fusion[J].,2018,28(02):152.[doi:10.3969/ j. issn.1673-629X.2018.08.032]
[3]郭蕾蕾,俞 璐,段国仑,等.基于 AP 聚类的多特征融合方法[J].计算机技术与发展,2019,29(08):47.[doi:10. 3969 / j. issn. 1673-629X. 2019. 08. 009]
　GUO Lei-lei,YU Lu,DUAN Guo-lun,et al.A Multi-feature Fusion Method Based on AP Clustering[J].,2019,29(02):47.[doi:10. 3969 / j. issn. 1673-629X. 2019. 08. 009]
[4]马文斌,夏国恩.基于深度神经网络的客户流失预测模型[J].计算机技术与发展,2019,29(09):76.[doi:10. 3969 / j. issn. 1673-629X. 2019. 09. 015]
　MA Wen-bin,XIA Guo-en.Customer Churn Prediction Model Based on Deep Neural Network[J].,2019,29(02):76.[doi:10. 3969 / j. issn. 1673-629X. 2019. 09. 015]
[5]安计勇,闫子骥.基于半监督学习的蛋白质相互作用预测模型[J].计算机技术与发展,2021,31(07):7.[doi:10. 3969 / j. issn. 1673-629X. 2021. 07. 002]
　AN Ji-yong,YAN Zi-ji.Prediction Model of Protein-protein Interactions Based on Semi-supervised Learning[J].,2021,31(02):7.[doi:10. 3969 / j. issn. 1673-629X. 2021. 07. 002]
[6]利向晴,夏国恩,张显全,等.基于深度神经网络权重集成的客户流失预测[J].计算机技术与发展,2021,31(10):18.[doi:10. 3969 / j. issn. 1673-629X. 2021. 10. 004]
　LI Xiang-qing,XIA Guo-en,ZHANG Xian-quan,et al.Customer Churn Prediction Based on Deep Neural Network Weight Ensemble[J].,2021,31(02):18.[doi:10. 3969 / j. issn. 1673-629X. 2021. 10. 004]

更新日期/Last Update: 2020-02-10