<p>The evaluation of the fatigue life of Ti–5Al–2.5Sn alloy reinforced with tungsten particles is crucial for guaranteeing durability in aerospace, automotive, and structural applications. This paper presents a data-driven approach for predicting fatigue life, utilizing Random Forest and Support Vector Regression (SVR). Data from experimental fatigue tests, conducted at different stress amplitudes and tungsten weight percentages, were utilized for training and validation purposes. The Random Forest model attained the highest predictive performance (R² = 0.824), closely succeeded by SVR (R² = 0.687). Each algorithm underscores the importance of stress amplitude, stress ratio, and tungsten content in determining fatigue life. This comparative analysis highlights the flexibility and precision of machine learning models, providing a significant resource for forecasting fatigue life beyond conventional empirical and finite element techniques.</p>