Customer Behaviour Analysis Through Clustering and Classification for Segmentation and Forecasting in Marketing Campaigns

Abstract

In this paper, the possibilities of extracting useful insights through data analysis using machine learning approaches are explored on a dataset focused on customer relationship management. Exploratory data analysis is applied to customers and their interactions, such as purchases, responses to marketing campaigns, and complaints. The detailed examination aims to assess the potential of using results from past marketing campaigns to predict customer reactions to future ones, thereby improving marketing effectiveness. To this end, a classification model is built to predict customer responses to the latest campaign. Evaluation metrics are calculated to assess the classification performance across different sets of selected features and classifiers, and the findings are summarized and discussed. Generalized Linear Model (GLM) and H2O Deep Learning (DL) models stood out as the best performers in the study.