Abstract
The availability of student data in educational institutions presents a valuable opportunity to apply machine learning (ML) techniques for predicting academic performance before the start of a new semester. Some variables can support early identification of students’ potential performance and guide them in planning. In this article, we present predictive analysis models for the academic performance of engineering students at Yarmouk University during the years between 2019 and 2024. The machine learning models were applied using data collected from the university’s registration office, with some demographic information. Various machine learning models were employed, including Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Decision Trees, and Artificial Neural Networks (ANN) to identify the significant attributes affecting student performance in the machine component design course. The results present that factors such as CGPA and registered course load (in credit hours) significantly affect student performance. Among the models applied, ANN achieved the highest accuracy with an RMSE of 6.1, and the model SVR the lowest RMSE of 8.2. This research aims to assist students in identifying the potential outcomes early and provide advice to support students in registration planning.
| Original language | English |
|---|---|
| Pages (from-to) | 1547-1555 |
| Number of pages | 9 |
| Journal | International Journal of Engineering Education |
| Volume | 41 |
| Issue number | 6 |
| State | Published - 2025 |
| Externally published | Yes |
UN SDGs
This output contributes to the following UN Sustainable Development Goals (SDGs)
-
SDG 4 Quality Education
Keywords
- AI in education
- academic performance prediction
- engineering students, student data analytics
Fingerprint
Dive into the research topics of 'AI-Based Performance and Grade Prediction for Undergraduate Industrial Engineering Students in Machine Component Design Course'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver