Evolutionary Model Based Student Performance Prediction: A Hybrid Approach

This research delves into the application of metaheuristics to evaluate student performance from various associated factors. By leveraging both academic and demographic data, this work aims at evaluating a student from various perspective using hybridized evolutionary algorithm. The standalone Particle Swarm Optimization (PSO) and Firefly Algorithm (FA) models, though effective, often suffer from premature convergence and suboptimal feature selection, particularly in high-dimensional data environments. To overcome these challenges, a PSO-FA model is proposed, combining PSO's strong global search capability with FA's refined local search, thereby achieving a more balanced exploration-exploitation trade-off. PSO-FA hybridization balances the limitations of both the algorithms. Experimental results reveal that the hybrid model outperforms traditional methods in terms of convergence and finding optimal solution. The hybrid approach demonstrates improved robustness, making it a promising tool for educational institutions aiming to enhance decision-making processes and student outcomes through data-driven insights adding a new direction to assessment procedure.