چكيده به لاتين
Nowaday, the emerging rapid growth of data volume on the one hand, and the increasing competence between the companies on the other hand, motivate the compaies to try to understand the importance of the information which could be resulted from their customers as well as to investigate different mechanisms in order to use such information. Understanding the behaviours of the customers involves the application of the methods and the techniques to be able to recognize and predict the future behavior and the properties of the different groups of the customers, as well as to enforce the companies’ appropriate sterategies base on such behavoiurs.
RFM is one of the best-known behavioural models which is extensively used in the customer segmentaion problem. This model inxorporates three (Recency, Frequency, Monetary) aspects of the customers’ behaviours. In this research study, we incorporated a new measuere as uncollected credit documents of the customers which is intended as a cost imposed by the customers. In fact, the so-called RFM is extended in order to give some beter in terms of customer behavior analysis criteria, since the identification of beneficial groups of the customers is a major requirements of the companies. This study incorporates K-means clustering for customer segmentation as well as Davies Bouldin index where the whole process of the segmentation is based on CRISP-DM methodology. The experimental results show that RFMC outperforms the competing models in terms of the quality of the customer’s clusters.
Keywords: Datamining ،CRM ،Clustering ،CRISP_DM ،K-means،K-medoid
