<![CDATA[Small and Medium Enterprises (SMEs) play an essential role in supporting economic growth, employment creation, and innovation across developing and emerging economies. To maintain competitiveness, SMEs need to improve their business performance by understanding the factors that influence organizational success. Several factors, including risk-taking, open innovation, cost leadership, proactiveness, aggressiveness, and autonomy, contribute to SME performance improvement. This study aims to develop and compare machine learning models for predicting SME business performance using these influencing factors. A dataset containing 283 SME records obtained from the Mendeley Data Repository was used in this study. Four machine learning approaches were evaluated, including Random Forest (RF), XGBoost, Artificial Neural Network (ANN), and ANN optimized using Particle Swarm Optimization (PSO). The models were assessed using regression performance metrics, including the coefficient of determination (R²), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and computational execution time. The experimental results indicate that RF achieved the best prediction performance with an R² of 0.924, RMSE of 0.120, and MAE of 0.065, demonstrating high predictive accuracy and low error. XGBoost and ANN-PSO also showed competitive performance, while ANN achieved moderate results. Therefore, RF is recommended as an effective model for SME performance prediction. Future studies should employ larger datasets and external validation to improve model generalizability.]]>
