Machine Learning Models in Customer Behaviour Prediction: A Comparative Analysis

Understanding and predicting customer shopping behaviour is essential for enhancing user experience, optimizing marketing strategies, and increasing sales in e-commerce. This study compares various machine learning models to forecast customer behaviour patterns using a dataset of customer demographics, purchase frequency, and product preferences. K-Means clustering identified distinct customer segments, which were utilized as the dependent variable for classification tasks. We evaluated K-Nearest Neighbours, Gradient Boosting, Support Vector Classifier, and Random Forest in terms of accuracy, precision, recall, and F1-score. These ML algorithms have been evaluated and their performance in customer behaviour prediction is compared. The feature importance analysis identified the main factors affecting customer behaviour, enabling more effective targeted marketing strategies. This paper emphasizes the importance of selecting appropriate models for behaviour prediction and offers a framework for e-commerce analytics. Comparative analysis of customer behaviour through different ML models has been evaluated on which, the Gradient Boosting and Random Forest models has given better performance in terms of accuracy and robustness. Our results indicate significant variations in model performance, with Random Forest and Gradient Boosting demonstrating superior accuracy and robustness compared to KNN and SVC.