Concerns regarding student retention at universities have persisted among educators for decades, primarily due to its effects on institutional rankings, reputation, and financial health. The College of Information and Computing Sciences (CICS) at Mindanao State University in Marawi City seemingly struggled with student retention for many years. This study aims to conduct a predictive analysis to identify the sociodemographic and academic factors that significantly influence a student’s choice to withdraw from their program and to visualize the results using Python in the Jupyter Notebook environment. The insights obtained could help decision-makers pinpoint areas requiring enhancement to reduce attrition and assist students in graduating on time. To determine the most suitable algorithm for the prediction, ten (10) machine learning models were considered: Artificial Neural Network (ANN), Decision Tree, Extreme Gradient Boosting (XGBoost), Gradient Boosting, K-Nearest Neighbors (KNN), Logistic Regression, Multi-Layer Perceptron (MLP), Naïve Bayes, Random Forest, and Support Vector Machine (SVM). These models were evaluated based on five (5) performance metrics: Accuracy score, F1 score, Mean Squared Error (MSE), Precision, and Recall. The results indicate that the Extreme Gradient Boosting (XGBoost) achieved the highest accuracy in predictions. Consequently, the model was selected as the most reliable method for accurately determining the likelihood of students either withdrawing from or remaining in the program.