Comparison of Machine Learning Methods in Customer Segment

Abstract

Customer segmentation plays a key strategy in marketing and business analytics. It assigns customers to different groups based on their common characteristics, which allows the company or organization to regulate their marketing efforts and product offers to meet specific needs of each group. With the development of machine learning, lots of methods are discovered and being used widely. The main purpose of this paper is to introduce the four popular machine learning methods and compare their functions. This paper first introduces the datasets in the field of customer segmentation, then it introduces customer segmentation methods based on machine learning, including Support Vector Machines (SVM), Decision Trees (DT), Random Forest (RF), and K-Nearest Neighbors (KNN). Based on the results, Random Forest offers the best precision which runs up to 89.61%. By introducing and comparing the performance of four different methods in a specific data environment, it will enlighten researchers in the field of customer segmentation.