点击次数：

论文类型：期刊论文

发表时间：2011-04-01

发表刊物：ICIC Express Letters

收录刊物：EI、Scopus

卷号：5

期号：4 B

页面范围：1411-1416

ISSN号：1881803X

摘要：This paper presents a hybrid clustering algorithm to be used for customer segmentation classifications in a mobile e-commerce environment. The proposed algorithm, named KSP, is based on three well know clustering algorithms: the K-means, SOM and PSO. We first applied those algorithms directly to a mobile customer segmentation problem using data collected from a well established restaurant chain which has operations throughout Japan. The results from the classifications were compared with the existing company customer segmentation data for verifications. Based on our initial analysis, special characteristics from those three algorithms were extracted and modified in our KSP method which performed extremely well with mobile e-commerce applications. Our result is 17.08% and 7.56% more accurate than that of K-means and SOM respectively. ICIC International ? 2011.