Work

Predicting Student Performance Through Machine Learning

Year: 2026
Type: article
Abstract: The rapid evolution of Educational Data Mining(EDM) has transformed how academic institutions addressstudent retention and success. This research focuses on thedevelopment of a predictive framework designed toidentify student performance outcomes by leveragingMachine Learning (ML) algorithms. By analyzing amultifaceted dataset—encompassing demographicattributes, socio-economic backgrounds, and historicalacademic records—the study evaluates the efficacy ofvarious supervised learning models, including RandomForest, Support Vector Machines (SVM), and GradientBoosting.Pre-processing techniques such as SMOTE for classbalancing and Recursive Feature Elimination (RFE) wereemployed to enhance model precision. Preliminaryresults indicate that ensemble methods significantly outperformtraditional classifiers, providing high accuracy inidentifying "at-risk" students before the conclusion of anacademic term. The findings offer a scalable solution foreducators toimplement data-driven intervention strategies,ultimately fostering a more personalized and proactiveeducational environment. This study underscores the potential of ML as a cornerstone for institutional decision-making and academic excellence
Language: English
Open Access status: hybrid
OpenAlex ID: W7161050390
FWCI:
Cites:
Cited by:
Related to: 10