Predicting Students’ Performance Using a Hybrid Machine Learning Approach
Year: 2025
Type: article
Abstract: Learning institutions have amassed volumes of educational data from sources such as learning management systems and academic databases. This presents an opportunity for learning institutions to predict students’ performance using Machine Learning (ML) techniques. This paper explores the application of a hybrid ML approach to students’ performance. Previous studies have employed individual ML algorithms for performance prediction; these models often suffer from limitations such as low accuracy and bias towards specific data characteristics. To address these challenges, this study proposes a hybrid ML approach that integrates multiple algorithms to leverage their strengths. The Random Forest classifier was used to reduce variance and handle nonlinear relationships. The Linear Support Vector Classifier (SVC) captured linear patterns within the data, and Logistic Regression was employed as a meta-learner to make the final predictions. By combining these classifiers, the proposed model aims to deliver improved prediction accuracy and robustness.
Source: Computing Open
Authors: Ropafadzo Duwati, Tawanda Mudawarima
Institution: Harare Institute of Technology
Language: English
Subfield: Computer Science Applications
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: diamond
OpenAlex ID: W4413249908