摘要
为精准有效地预测电商客户粘性,提升经济效果,研究数据挖掘技术在电商客户粘性预测中的应用。利用可变网格的K-means聚类算法,聚类获取电商客户粘性预测相关数据;通过有效性指标优化可变网格K-means聚类算法的聚类数,确定最佳聚类数,提升数据聚类效果;采用技术接受模型,依据聚类获取的数据,建立电商客户粘性预测指标体系;通过模糊层次分析法,结合指标体系,建立电商客户粘性预测模型,获取预测分值。实验结果表明,该模型可有效确定最优聚类数,精准聚类电商客户粘性预测相关数据;所建立的预测指标体系的指标相关性较低,结构较稀疏、较全面。总体说明,该模型可有效预测电商客户粘性。
In order to accurately and effectively predict e-commerce customer stickiness and improve economic results, the application of data mining technology in e-commerce customer stickiness prediction is studied. The K-means clustering algorithm of variable grid was used to obtain the relevant data of e-commerce customer stickiness prediction. The clustering number of variable grid K-means clustering algorithm was optimized by the validity index to determine the optimal clustering number and improve the data clustering effect. The technology acceptance model is adopted to establish the prediction index system of e-commerce customer stickiness according to the data obtained by clustering. Through the fuzzy analytic hierarchy process, combined with the index system, the e-commerce customer viscosity prediction model is established to obtain the prediction score. The experimental results show that the model can effectively determine the optimal clustering number and accurately cluster the relevant data of ecommerce customer stickiness prediction. The index correlation of the established prediction index system is low, and the index system structure is sparse and comprehensive. Overall, the model can effectively predict e-commerce customer stickiness.
作者
黄维雅
HUANG Wei-ya(School of Economics and Trade,Xiamen Xingcai Vocational and Technical College,Fujian Xiamen 361000,China)
出处
《齐齐哈尔大学学报（自然科学版）》
2023年第1期81-86,94,共7页
Journal of Qiqihar University(Natural Science Edition)
基金
2020年度福建省教育厅中青年教师教育科研项目“大数据分析在网络平台精准营销中的应用研究”(JAS20752)。
关键词
数据挖掘技术
电商客户
粘性预测
可变网格
K-MEANS聚类
模糊层次分析
data mining technology
e-commerce customers
viscosity prediction
variable grid
K-means clustering
fuzzy hierarchy analysis
