Machine Learning Techniques and Recommender Systems for Large Educational Data
E-learning approaches allow learners with a range of courses on online platforms to choose appropriate courses according to their preferences and interests. Recommender systems show a role in helping students select courses via analyzing their data and understanding interests. In the domain of e-lea...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
University of Anbar
2024-12-01
|
| Series: | مجلة جامعة الانبار للعلوم الصرفة |
| Subjects: | |
| Online Access: | https://juaps.uoanbar.edu.iq/article_185697_6aed795e372fa7d290b5ed958114a872.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | E-learning approaches allow learners with a range of courses on online platforms to choose appropriate courses according to their preferences and interests. Recommender systems show a role in helping students select courses via analyzing their data and understanding interests. In the domain of e-learning platforms, there is ability to improve recommendation systems utilizing machine intelligence (MI) techniques. This study aims to proposed recommendations for e-learning courses based on user assessments so to achieve this aim, we need to apply a collaborative filtering technique. This study utilized the educational dataset from Coursera’s online courses. This study proposed machine learning (ML) models such as K Nearest Neighbor (KNN) and Singular Value Decomposition (SVD) while this paper utilizing evaluations of performance metrics like root mean square error (RMSE), mean absolute error (MAE), hit rate (HR), average reciprocal hit rating (ARHR), accuracy, and recall. The finding results show the SVD achieved a 91% HR and KNN achieved a 96% HR while the KNN algorithm achieved highest accuracy of 99%, closely followed by SVD at an accuracy of 96% but when utilized the Grid Search (GS) technique with the recommended ML algorithms achieved HR of 99% for SVD and 98% for KNN. Although the accuracy only showed slight improvements after the changes, these results highlight the efficacy of collaborative filtering methods in improving recommendations in e-learning settings, thereby enhancing the customized learning experiences of users. |
|---|---|
| ISSN: | 1991-8941 2706-6703 |