期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2016
卷号：84
期号：2
出版社：Journal of Theoretical and Applied
摘要：The number of existing competition between companies make a company does not only focus on product development, but also in relation to customers. Customer Relationship Management (CRM) is a strategy to manage and maintain relationships with customers as well as an attempt to determine the wants and needs of customers. The analysis needs to produce the information on which customers are valuable or not. This will allow companies to implement efficient and effective measures to be loyal customers. This research is a data mining process with the method used is clustering by Affinity Propagation and Recency Frequency and Monetary (RFM) model on 1.000 Customer data. Distance method used on Affinity Propagation algorithm is Euclidean Distance and Manhattan Distance. Application was built using MATLAB to facilitate companies to analyze customer transaction data. From 1.000 customer data it generates 51 clusters. The best results if used Euclidean distance with preference median or Manhattan distance with preference minimum. The results of the clustering process is analyzed using comparison of RFM model and divide customer into eight type of customer and then into three groups. The results of customer segmentation are Most Valuable Customer 3.7%, Most Growable Customer 26.3% and Below Zero Customer 70%.
