Customer Segmentation Using Two-Step Mining Method Based on RFM Model

Many companies are unable to maximize the use of information that would strengthen the company competitive market such as, the use of Customer Relationship Management (CRM) to gain insights into customer demands and needs. In addition, even though combination of mining user data and computational methods have been used, those approaches have some limitations. Also, only a few studies use RFM model in understanding the user. Furthermore, data mining techniques applied in the business environment is not yet descriptive. To overcome those problems, a two-step mining method is implemented in this study. Recency Frequency Monetary (RFM) model is used as the basis to create customer segmentation based on customer Recency, Frequency, and Monetary. The first step, the data is extracted to RFM model and then clustered using k-Means algorithm. In the second step, the data in each cluster is analyzed using association analysis to create customer characteristics represented by IF-THEN rules. The cluster results are analyzed using Silhouette and Connectivity measure. The rules are used to mine frequent patterns in each cluster and to explain each cluster to be more descriptive. The results show that the information in each cluster of customer segmentation has more valuable information that may benefit in understanding the customer for the company to apply specific strategies to increase company profits.