Machine Learning Approach for Customer Segmentation and Prediction: The Case of Oromia Saving and Credit Shared Company

Abstract

Identifying customers which are more likely potential to have product and service offers are an important issue. In customers’ identification, the machine learning approach has been used extensively to segment and predict potential customers for a product and service.The aim of this study is to create a model that helps to classify customers for Oromia Credit and Saving Share Company microfinance institution products and services. Since there are not any predefined classes, that describe the purchasers of the institution, the researcher uses clustering techniques that resulted within the appropriate number of clusters. Then, a predictive model was developed to predict the potential of the purchasers. This predictive model achieved an accuracy of 94.1%. For modeling purposes, data was gathered from an establishment head office. Since irrelevant features end in bad model performance, data preprocessing was performed so as to work out the inputs to the model.Thus, various data processing techniques and algorithms were wont to implement each step of the modeling process and alleviate related difficulties. K-means was used as a clustering algorithm to segment customers’ records into clusters with similar characters. Different parameters were wont to run the clustering algorithm before reaching a segment that made business sense. The J48 decision tree algorithm was used for prediction purposes. Additionally to those attributes that are believed by the experts to possess a high impact on customer segmentation, attributes value of loan amount features a big influence.

Keywords: Microfinance, machine-learning, predictive analytics, segmentation, data mining

DOI: 10.7176/CEIS/13-2-03

Publication date:March 31^st 2022