Scrutiny of The Results Obtained from The Unsupervised Machine Learning Algorithms for The Customer Segmentation Problem

Authors

  • Reetu Jain

Abstract

Customer segmentation (CS) is a strategy to focus on the group of similar spending behavior customers. The advantage of CS is to closely align their policy and tactics to better target the customers. Segmenting customers on the basis of their spending behavior can be done with the help of machine learning as it is a great tool for analyzing and finding patterns in the dataset. The study is conducted in two phases. The first phase includes an exploratory data analysis (EDA) that helps in understanding the customer’s traits and their spending behavior. The second phase of the study includes the development of the ML model. For the problem of CS, there are no label outputs for which the data are trained. Hence unsupervised learning (UL) is the most preferred ML tool for the CS problem. Groupings of the customers are computed using two UL models namely k-means and density-based spatial clustering of applications with noise (DBSCAN). The former model grouped the data into 5 clusters whereas the later model grouped it into 3 clusters along with some data identified as noise. The noise from the context of the problem of CS refers to those customers for who cannot be placed in any clusters. The application of UL in this study may further open up the potential for other applications in the same industry.

Downloads

Published

2022-06-18

How to Cite

Jain, R. (2022). Scrutiny of The Results Obtained from The Unsupervised Machine Learning Algorithms for The Customer Segmentation Problem. iJournals:International Journal of Software & Hardware Research in Engineering ISSN:2347-4890, 10(6). Retrieved from https://ijournals.in/journal/index.php/ijshre/article/view/135

Most read articles by the same author(s)

1 2 > >>