重庆理工大学学报2024，Vol.38Issue(20)：138-149,12.DOI:10.3969/j.issn.1674-8425(s).2024.10.011
基于RFM的聚类算法在零售市场客户细分研究
Research on customer segmentation in retail market based on RFM clustering algorithm
摘要
Abstract
Customer relationship management as an important part of enterprise management,its customer segmentation function directly affects the enterprise marketing strategy.In order to better segment the retail market customers,the RFM model and four clustering algorithms including K-means,DBSCAN,AGNES and GMM are verified in UCI Online Retail by applying a British retailer data set.The results of customer classification on retail retailer data set are compared with those of the above four clustering algorithms by using profile coefficient,Kalinsky Harabas Index(CHI)and Davidson Burger Index(DBI).The empirical results show that K-means and AGNES algorithm have better clustering effect on the selected retailer data set,while DBSCAN and GMM algorithm have less clustering effect,aiming to provide reference and reference for Machine Learning Clustering Algorithm in customer classification based on RFM model.It is recommended that enterprises attach importance to the output data,improving the system related to enterprise data,and combining the characteristics of customer data with the sales characteristics of the enterprise itself to use clustering algorithms for targeted customer segmentation,assisting in summarizing customer profiles,and thus developing targeted marketing strategies.关键词
客户细分/机器学习/RFM/聚类算法/零售市场Key words
customer segmentation/machine learning/RFM/clustering algorithm/retail market分类
管理科学引用本文复制引用
吴花平,冯薇薇,李林..基于RFM的聚类算法在零售市场客户细分研究[J].重庆理工大学学报,2024,38(20):138-149,12.基金项目
教育部人文社会科学研究规划基金项目"智慧审计驱动更高水平全民健身公共服务绩效提升研究"(23YJA890039) （23YJA890039）
重庆理工大学研究生教育高质量发展项目"企业网络安全与企业高质量发展——基于文本分析和机器学习的经验证据"(gzlcx20233459) （gzlcx20233459）
