Customer Segmentation Using K Means Algorithm and RFM Model

Customer segmentation and pattern extraction are two of the most crucial features of a business decision support. To expand the business successfully in a cutthroat industry as it is very important to identify the potential of clients as fast as possible. Managers of the organization make wise strategic decisions after identifying the targeted customers and their corresponding purchase patterns. This paper provides an integrated innovative technique for defining target clients and providing to them to increase the organization's revenue. The major objective of this paper is to classify the data of clients into groups on the basis of their characteristics by using the K-means algorithm. RFM model is used to rank and group to customers on the basis of their recent transactions. A new client is assigned to the cluster by taking a mean value as primary indicator. Elbow method is used for selecting the number of clusters initially and Davies-Bouldin score is used to determine optimal number of clusters. Flattened graph and Snake plot for different K values are shown for analyzing the clusters and finalizing the K value.