Abstract
This paper proposes the recommendation system which is advanced using RFM method and Association Rules in e-Commerce. Using a implicit method which is not used user's profile for rating, it is necessary for user to keep the RFM score and Association Rules about users and items based on the whole purchased data in order to recommend the items. This proposing system is possible to advance recommendation system using RFM method and Association Rules for cross-selling, and also this system can avoid the duplicated recommendation by the cross comparison with having recommended items before. And also, it's efficient for them to build the strategy for marketing and crm(customer relationship management). It can be improved and evaluated according to the criteria of logicality through the experiment with dataset collected in a cosmetic cyber shopping mall. Finally, it is able to realize the personalized recommendation system for one to one web marketing in e-Commerce.
이 논문은 RFM 기법과 연관성 분석을 이용한 개인화된 전자상거래 추천 시스템을 제안한다. 제안된 전자상거래 추천시스템은 사용자의 평가 자료에 의존하지 않고 묵시적인(Implicity)방법을 이용하여 고객정보와 구매이력 정보를 기반으로 RFM(Recency, Frequency, Monetary) 기법을 이용한 고객 세분화와 교차판매(cross-sell)관계를 찾는 연관성 분석을 이용한 개선된 시스템이다. 또한 고객군별 구매특성 분석을 통하여 효율적인 마케팅 전략과 고객관계관리(CRM: Customer Relationship Management)방법을 제시한다. 현업에서 사용하는 데이터 셋을 구성하여 실험 및 평가를 통해서 효용성을 입증 및 평가하여 일대일 웹 마케팅을 실현하였다.
