Abstract: The stability of unsaturated slopes, as one of the important issues in geotechnical engineering, plays a fundamental role in evaluating the safety of civil engineering projects and in preventing occurrence of instability under the influence of hydrological and geometrical parameters. In this study, in order to evaluate the stability of unsaturated slopes, numerical modeling based on the finite element method was performed using the PLAXIS 2D software, and based on its results, a dataset corresponding to stable and unstable conditions was extracted. Subsequently, the performance of four machine learning algorithms, including Support Vector Machine, Logistic Regression, Random Forest, and XGBoost, was evaluated on these data. The implementation and coding of these models was done using the Python programming language. The results indicated that all the investigated models demonstrated acceptable performance and very close results in classifying slope stability conditions. The analysis of feature importance and the effects of input parameters showed that groundwater level, slope angle, rainfall intensity, and rainfall duration all have an increasing effect on slope instability. An increase in these parameters leads to a reduction in the factor of safety and an increase in the probability of instability occurrence. From a soil mechanics perspective, this behavior is attributed to the reduction of matric suction, the decrease in effective stress, and the reduction in shear strength due to water infiltration, as well as the increase in the driving shear stress component in steeper slopes.