This study investigates the efficacy of Naive Bayesian algorithms combined with a rule-based classifier to predict and map the academic performance of Bachelor of Science in Business Administration major in Marketing Management (BSBAMM) students at the Polytechnic University of the Philippines (PUP). Amidst disruptions caused by the COVID-19 pandemic, data-driven approaches are increasingly vital in higher education to enhance student outcomes. Historical data encompassing student demographics and academic records were analyzed to develop a predictive model, achieving 95% accuracy in forecasting student performance. This research underscores the potential of machine learning in identifying at-risk students early, facilitating timely interventions and personalized learning paths. The findings contribute valuable insights for educators and institutions seeking to optimize resource allocation and improve graduation rates.

Keywords: Predicting academic performance, higher education students, Naive Bayesian algorithm, Rule-based algorithm, historical data, student demographics, student success, student failure, machine learning algorithms, resource allocation, timely interventions, student outcomes