Ensemble-Based Prediction Model for Enhanced Electronics Engineering Licensure Examination Results Using Student Performance Analysis

This research focuses on analyzing the impact of Electronics Engineering graduates' performance in licensure examinations on the overall quality of Higher Education Institutions (HEIs). Data mining techniques, specifically classification, are employed to examine patterns within academic performance data and predict the success of graduates in licensure examinations. The findings aim to enhance the understanding of factors influencing licensure examination performance and guide institutions in improving their instructional strategies and curriculum design. The research aims to improve prediction accuracy by training and testing different models and optimizing their hyperparameters. The ensemble model emerges as the best-performing model, combining the predictions of multiple models and achieving high accuracy rates of 95.33% and 98% in the 70:30 and 80:20 data splitting ratios, respectively. This research contributes to the application of data mining techniques in educational settings, particularly in analyzing Electronics Engineering licensure examination performance. It provides valuable insights for HEIs in evaluating and enhancing their educational programs to better prepare students for professional practice.