陈俊龙 吴丽丽



摘 要:为进一步探究和分析电子商务客户关系,本文提出e价值的指标体系和计算方法,同时基于使用k-means方法对客户进行分类,实现对客户关系的深层发掘。基于改进的RFM模型实现了对客户的辨别与分类功能,对不同客户的e价值能进行有效预测,同时可以为企业在电商相关领域营销策略的差异化实施提供依据。对客户关系进行深层细分。同时基于AdaBoost分类器,提出以C5.0决策树作为基分类器的客户保持与流失预测模型,降低错误预测成本,精准识别高价值客户。
关键词:RFM;AdaBoost;电子商务;客户价值
中图分类号:TP391. 41 文献标识码:A DOI:10.3969/j.issn.1003-6970.2021.03.001
本文著录格式:陈俊龙,吴丽丽.基于RFME模型和AdaBoost分类器的电子商务客户关系研究[J].软件,2021,42(03):001-007
Research on E-commerce Customer Relationship Based on RFME Model and AdaBoost Classifier
CHEN Junlong, WU Lili
(College of Information Science and Technology, Gansu Agricultural University, Lanzhou Gansu 730070)
【Abstract】:In order to further explore and analyze the relationship between e-commerce customers, this article proposes an index system and calculation method for e-value, and at the same time classifies customers based on the use of k-means method to realize in-depth exploration of customer relationships. Based on the improved RFM model, the function of identifying and categorizing customers is realized, and the e-value of different customers can be effectively predicted. At the same time, it can provide a basis for the differentiated implementation of marketing strategies for companies in the e-commerce-related fields. In-depth segmentation of customer relationships. At the same time, based on the AdaBoost classifier, a customer retention and churn prediction model based on the C5.0 decision tree is proposed to reduce the cost of error prediction and accurately identify high-value customers.
【Key words】:RFM;AdaBoost;E-commerce;customer relationship management
0 引言
在网络技术日新月异的当下,电子商务平台已经深入各行各业中,生活中处处可见电商领域的产品或服务。在社会发展和进步的同时,电子商务区别于以往传统的消费模式,作为全新的形式冲击全国受众的普遍认知,并使其购买行为产生了或多或少的变化。据资料显示,截止2015年,我国互联网用户已逾6亿,到2019年6月,我国网民规模升至8.54亿,手机网民规模达8.47亿,网络普及率超过61.2%。至2020年3月,我国互联网普及率已达到64.5%,网络购物用户规模达7.10亿,近2015年的两倍[1]。
2019年,我国互联网交易规模达10.63万亿元,与繁荣发展的互联网业态相对应,在全新商务模式下,对消费者管理和客户价值认知评估模式也需要推动发展与转型。与传统行业的销售模式相比,在电子商务环境下,消费者的选择空间极度扩大,信息流动迅速,客户留存率大大降低。……