This study addresses the phenomenon of academic dropout that affects higher education institutions and the achievement of educational quality standards. In this context, factors related to university students are analyzed, such as their integration into the institution and their performance during the early academic semesters, based on historical institutional, academic, socioeconomic, and personal data. The objective of this study is to apply these factors to classification models in Machine Learning to predict academic dropout in Engineering programs at the Universidad Nacional de Moquegua (UNAM). For modeling, Decision Tree, Random Forest, Simple Perceptron, and Logistic Regression models were applied. The performance of the algorithms was evaluated using the confusion matrix and other performance metrics. The results revealed that the Decision Tree algorithm performed the best, achieving an accuracy of 99% and an F1-score of 98%. Additionally, the processed data identified the most relevant factors for predicting academic dropout, such as the student’s current status, the number of courses passed, age, and the parents’ educational level. These findings enable the deployment of the algorithm as a useful tool for the departments responsible for student academic monitoring.