COMPARATIVE PERFORMANCE ANALYSIS OF MACHINE LEARNING ALGORITHMS FOR BUSINESS INTELLIGENCE: A STUDY ON CLASSIFICATION AND REGRESSION MODELS

This study presents a comparative analysis of five widely used machine learning algorithms—Logistic Regression, Support Vector Machines (SVM), Random Forest, Gradient Boosting, and Neural Networks—in the context of business intelligence (BI). The performance of these models was evaluated on both classification and regression tasks, utilizing a comprehensive set of metrics including accuracy, precision, recall, F1 score, AUC-ROC for classification, and R-squared for regression. Results indicate that ensemble models, particularly Random Forest and Gradient Boosting, outperformed other algorithms across both tasks. Random Forest achieved the highest AUC-ROC (96.3%) in classification, while Gradient Boosting led with the highest F1 score (94.2%) and AUC-ROC (97.8%), reflecting its ability to model complex, non-linear relationships. In regression tasks, Gradient Boosting (R² = 0.94) and Random Forest (R² = 0.91) demonstrated superior explanatory power. While Neural Networks (R² = 0.93) performed well, their computational complexity and lack of interpretability pose challenges for certain BI applications. Logistic Regression and SVM, though effective in simpler contexts, were generally outperformed by more complex models. The findings emphasize the importance of selecting the appropriate model based on the business objectives, data characteristics, and computational resources, with ensemble methods being ideal for high-accuracy, complex BI tasks. This study contributes valuable insights for organizations aiming to leverage machine learning for data-driven decision-making and enhances the understanding of algorithmic trade-offs in business intelligence.