Examination of the Criticality of Customer Segmentation Using Unsupervised Learning Methods

In the world, everything revolves around selling and buying to get something or to earn a living. Whoever is selling is a seller who needs a customer to sell the things. The customer went to a seller when the seller approached the customer. Long-term relationships with customers become more and more important as a marketing paradigm unfolds. To predict the customer–seller relationship or to analyze customer satisfaction, to efficiently identify and serve its customers depending on multiple variables, a corporation must segment its market because it has a finite number of resources. Clustering is a useful and popular method for market segmentation, which identifies the intended market and customer groupings, in the field of market research. This study demonstrates how to segment mall customers using machine learning methods. This is the unsupervised clustering problem, and three well-known algorithms—K-means, affinity propagation, and DBSCAN—will be discussed and contrasted. The primary goal of the study is to go through the fundamentals of clustering techniques while also touching on some more complicated ideas. The study also revealed that there are more female customers than male consumers, with women making up 56% of all customers. Males have a greater mean income than females ($62.2 k vs. $59.2 k). Additionally, male customers’ median income ($62.5 k) is higher than female customers ($60 k). Both groups’ standard deviations are comparable. With an annual income of roughly 140 k dollars, one male stands out in the group.