Date of Publication :25th February 2025

Abstract: This paper investigates and compares the performances of three machine-learning models: Random Forest, XGBoost, and the Multi-Layer Perceptron, which is often referred to as MLP. As regards evaluation criteria, the mean performances were assessed in terms of f1-score, precision, and recall. The results show that, although Random Forest and XGBoost achieved almost perfect precision and recall, the MLP achieved a higher overall f1-score, indicating a superior balance between precision and recall. This comparison brings forth the tradeoffs of model selection in classification; that is, MLP is best for balanced performance, whereas Random Forest and XGBoost excel if false positives and false negatives are to be minimized.

Reference :