Valued for its nutritional and economic value, olive oil (OO) has been subject to adulteration practices. This study aimed to develop a classification model based on chemical composition to identify OOs from three main cultivars: Kalinjot, Ulli i bardhë Tirana, and Mixan, which comprise the majority of plantations in Albania. Eighty-five OO samples spanning different crop years and locations were studied using three different machine-learning algorithm models. The performance metrics of the k-Nearest Neighbors (kNN), Support Vector Machine (SVM), and Random Forest are discussed in terms of their classification performance. The comparison of accuracy revealed that the Random Forest model outperformed the others, achieving an accuracy of approximately 93%, compared to 81% for kNN and 78% for SVM. This significant finding, along with the clear confusion matrix of Random Forest, selects it as the preferred model for distinguishing OO based on cultivar and geographic origin. This project will help oil and oil extraction companies verify the authenticity of their products and detect adulteration practices in the Albanian oil sector.