<![CDATA[The rapid digital transformation in the retail sector has generated massive volumes of consumer transaction data stored within retail information systems. Although these data hold strategic value for decision-making, their utilization often remains limited to descriptive reporting. This study aims to analyze and predict consumer purchasing behavior by integrating machine learning–based predictive analytics into retail information systems using the Kaggle retail transaction dataset. The research methodology includes data preprocessing, exploratory data analysis, feature selection, and predictive model development using logistic regression, decision tree, and random forest algorithms. Model performance was evaluated using accuracy, precision, recall, and ROC–AUC metrics. The results indicate that the random forest model outperformed the other algorithms, achieving an accuracy of 88.76%, precision of 87.92%, and recall of 86.48%, demonstrating superior discriminative capability. These findings confirm that ensemble-based learning methods effectively capture complex and non-linear consumer purchasing patterns. The study contributes theoretically by extending the role of retail information systems from descriptive reporting tools to predictive decision-support systems, while practically providing a robust analytical framework to support inventory optimization, targeted promotion strategies, and personalized service delivery in data-driven retail environments.]]>
Copyrights © 2026
