The main objective of this research is to construct landslide susceptibility maps that are based not only on a set of geomorphological, geological, and hydrological factors but also on geotechnical parameters that are directly related to the shear strength of the ground, such as the Plasticity Index, Clay Fraction and Geological Strength Index. These parameters are overlooked in landslide susceptibility studies but are crucial for a comprehensive understanding of slope instabilities. Three methods were utilized to develop the appropriate classifiers: Logistic Regression, Random Forest and XGBoost. A database of 2500 landslide points and 2500 no-landslide points was selected for the study area, which is the southwestern part of Cyprus. These were divided into the training (70% of the total data) and a validation (30% of the total data) datasets. After processing the feature importance of 17 causal factors, lithology emerged as the most influential factor, followed by slope angle and land use. The evaluated geotechnical factors are highly important, as two of them rank sixth and seventh in the importance hierarchy. The training and testing capabilities of the three models were validated and compared with the results of the ROC curve analysis and estimation of several statistical metrics. Generally, all the models achieved high values of accuracy, sensitivity, and Cohen’s Kappa Index. Regarding the predictive capability, it was found that all the methods achieved very high accuracies, as they succeeded in identifying the 87.0% (LR), 96.3% (XGB), and 98.7% (RF) of the recorded landslides as areas of high to very high landslide susceptibility.