摘要
针对传统协同过滤算法在数据来源上依赖客户主观评价,以及现有推荐系统中缺少个性化推荐形式的缺陷,提出利用RFM模型在反映客户购买偏好和客户价值方面的良好表征性,将RFM模型与原协同过滤机制进行结合,对传统的协同过滤算法进行了改进,并制定了差异化的电子商务推荐策略,使推荐方式更符合不同客户的个性化,从而实现推荐内容和推荐形式的全面个性化.
Considering the deficiency in the traditional collaborative filtering algorithm whose original data were from customers' subjective evaluation and the lack of personalized recommendation pattern of current recommender systems,this paper combined RFM model with traditional collaborative filtering mechanism to improve the traditional collaborative filtering.Meanwhile,it drew up discrepant recommendation methods.This method can reflect customers' purchase preference and customers' value by taking advantage of the characteristics in RFM.Also,it can provide personalized services for customers in all-round manner.
出处
《江苏科技大学学报(自然科学版)》
CAS
北大核心
2010年第3期285-289,共5页
Journal of Jiangsu University of Science and Technology:Natural Science Edition
关键词
电子商务
推荐机制
协同过滤
RFM模型
electronic commerce
recommendation mechanism
collaborative filtering
RFM model