International Journal of Scientific Research in Engineering and Management (IJSREM) Volume: 05 Issue: 07 | July - 2021 ISSN: 2582-3930 © 2021, IJSREM | www.ijsrem.com | Page 1 A Study on Machine Learning Approach for Market Segmentation [Mr. Prateek Dutta 1 ] Bachelors of Technology in Artificial Intelligence, India Abstract Customer segregation can be a powerful way to identify unsatisfactory customer needs. This approach can be used by companies to outperform the competition by building more attractive products and services. Customer profile and purchase history were treated as internal data while server log, cookies and survey data were considered as external data. This information can be processed using one of several methods: Business Rule, Magento, Customer Performance, Quantile Membership, RFM Cell Separation Collection, Supervised Collection, Customer Identity Collection, Merge Purchase and Unchecked Mergers. Big data concepts and machine learning have encouraged greater acceptance of automated customer segregation methods by harvesting traditional market statistics that do not work whenever the customer base is too large. In this paper, the k-means clustering algorithm is used for this purpose. The Sklearn library is designed with the k- Means algorithm and the database available for commercial use. Features the average customer purchase number and the monthly customer average number. Keyword: Clustering, Customer Analytics, K-means, Python, Segmentation, Data Mining 1. Introduction The development of ecommerce began as the internet grew and continues to this day, especially in B2C ecommerce (Business to Customer). When shopping using ecommerce, the user finds it easier and faster [1]. Excessive information can be overcome by implementing personalization in ecommerce services such as providing product recommendations, linking recommendations, ads or text and graphics tailored to the user's features and needs [2]. In addition to solving the problem of over-information, customized services in ecommerce can maintain customer loyalty of existing customers, gain new customers by providing services to customers according to their needs and features [3]. Customer segregation is market segregation by different customer groups sharing the same characteristics. Customer segregation can be a powerful way to identify unsatisfactory customer needs. Using the above data companies can bypass the competition by creating more attractive products and services. Intelligence Segmentation Intelligence to improve marketing by providing products or services that meet the needs of each customer group. Collica-segregation is a process of classifying or classifying an object into a group with a similar feature and in the CRM (Customer Relationship Management) category used to classify customers according to a certain similarity by separating customer database records [4]. Boone and Roehm [5] studied Hopfield-Kagmar (HK) clustering method of customer segmentation using Hopfield’s artificial neural network technology. The study has shown that, each neuron in HK clustering method is connected with other neurons , and information can flow between neurons in multiple directions, which is more suitable for customer segmentation than the K-means clustering method; Kim et al. [6] used neural network clustering method to segment the customers of tourism; contrasting K-means, self- organizing map neural network and particle swarm optimization for three kinds of clustering algorithm, Deng et al. proposed hybrid clustering algorithm which was used for segmentation problem of catering industry customer. The ways in which businesses segments their customers information are: