改进的RFM模型和K-means算法在会员分类中的应用研究
Research on the Application of Improved RFM Model and K-Means Algorithm in Membership Classification
张利斌1
作者信息
- 1. 常州信息职业技术学院智能装备学院 江苏常州 213164
- 折叠
摘要
针对传统RFM模型用于会员分类会产生失真的问题,对RFM模型提出了改进,增加了客户关系长度和客户购买周期两个参数.同时针对传统的K-means算法存在的问题,提出了一种基于样本对象特征方差加权与中心初始化的K-means算法.利用改进的RFM模型对会员进行分类,可以有效地提高分类效率.
Abstract
Aiming at the distortion caused by traditional RFM models used for member classification,this paper proposes an improve-ment to the RFM model by adding two parameters:customer relationship length and customer purchase cycle.At the same time,a K-means algorithm based on sample object feature weighting and central initialization is proposed to address the prob-lems of traditional K-means algorithms.Using the improved RFM model for member classification can effectively improve classification efficiency.
关键词
RFM模型/K-means聚类/会员分类Key words
RFM model/K-means clustering/membership classification引用本文复制引用
出版年
2024
NETL
NSTL
万方数据