Enhancing Graduate Academic Performance Prediction and Classification: An Analysis Using the Enhanced Correlated Feature Set Model

Abstract

Complex assessment processes and limited improvement opportunities contribute to the challenges currently confronting higher education institutions. Recent focus shifts in academic research have sought to leverage the unique datasets generated within these institutions, aiming to refine student identification and performance metrics. This study primarily investigates the role data analysis and mining play in augmenting the assessment capabilities within these education institutions. The uncertainties surrounding student progression and retention rates necessitate the systematic evaluation of extensive data amassed within institutional databases and Learning Supervision Method (LSM) datasets. The intent is to provide academic professionals with detailed analytics on student activities and progression, enabling the tailoring of support to individual students and thus potentially increasing the efficiency of their coursework completion. In response to the growing urgency to evaluate and improve graduation outcomes without compromising educational quality, this study employs classification and data mining techniques to analyze graduate student academic performance. Educational institutions, striving to optimize limited resources, stand to benefit from the recognized decision-making power of deep learning models. A comparative analysis of the performance of elementary school students in college settings versus college students revealed a superior performance by the elementary school cohort. The proposed Graduate Interlinked Precedent Academic-based Performance Analysis using Enhanced Correlated Feature Set Model (GIPA-PA-ECFSM) illuminated a significant correlation between third-year and final-year performances. The proposed model aids in the identification of student performance across semesters, facilitating more effective student monitoring. When compared with traditional models, the proposed model demonstrates superior performance in analyzing graduate performance levels. 2023 IIETA.