Weighted Mining Association Rules Based Quantity Item with RFM Score for Personalized u-Commerce Recommendation System

摘要

This paper proposes a new weighted mining technique based quantity item with RFM(Recency, Frequency, Monetary) score for personalized u-commerce recommendation system under ubiquitous computing, or pervasive computing environment. Traditional association rule mining ignores the difference among the transactions. In this paper, it is necessary for us to consider the quantity of purchased data by each rank of RFM score in order to have different weights for different transactions, to generate weighted association rules through weighted mining association rules based quantity item with RFM score, and to recommend the items with high purchasability according to the threshold for creative weighted association rules with w-support, w-confidence and w-lift. To verify improved performance, we make experiments with dataset collected in a cosmetic internet shopping mall.