Customer Segmentation using K-means Clustering

Abstract

The zeitgeist of modern era is innovation, where everyone is embroiled into competition to be better than others. Today's business run on the basis of such innovation having ability to enthral the customers with the products, but with such a large raft of products leave the customers confounded, what to buy and what to not and also the companies are nonplussed about what section of customers to target to sell their products. This is where machine learning comes into play, various algorithms are applied for unravelling the hidden patterns in the data for better decision making for the future. This elude concept of which segment to target is made unequivocal by applying segmentation. The process of segmenting the customers with similar behaviours into the same segment and with different patterns into different segments is called customer segmentation. In this paper, 3 different clustering algorithms (k-Means, Agglomerative, and Meanshift) are been implemented to segment the customers and finally compare the results of clusters obtained from the algorithms. A python program has been developed and the program is been trained by applying standard scaler onto a dataset having two features of 200 training sample taken from local retail shop. Both the features are the mean of the amount of shopping by customers and average of the customer's visit into the shop annually. By applying clustering, 5 segments of cluster have been formed labelled as Careless, Careful, Standard,Target and Sensible customers. However, two new clusters emerged on applying mean shift clustering labelled as High buyers and frequent visitors and High buyers and occasional visitors.