Predicting student performance has been a critical focus for educational institutions seeking to enhance academic outcomes and support students in achieving their potential. This paper presented a mathematical framework for student performance prediction using Particle Swarm Optimization (PSO). The proposed model utilized key academic data, including midterm and final exam scores, assignment grades, attendance records, cumulative GPA, classroom environment (such as teacher-student ratio and available resources), and teaching methods (ranging from traditional to technology-enhanced approaches). By optimizing these variables using PSO, the framework identified patterns and relationships that significantly influenced student success. The application of PSO enabled efficient exploration of the feature space, providing accurate predictions while addressing challenges of non-linearity in the data. The framework’s predictive power offered valuable insights for educators and administrators, enabling data-driven interventions to improve student learning outcomes. Initial experiments yielded promising results, establishing the model's potential for broader use in academic performance forecasting.

Student performance prediction Particle Swarm Optimization Mathematical model educational data mining Optimization algorithm