The present paper proposes an e-learning system that combines popularity and collaborative filtering techniques to recommend pedagogical resources. A recommender system helps users get a correct and personalized decision by applying several recommendation methods such as content-based, collaborative filtering, and other hybrid approaches. However, predicting a relevant resource with a specific context, like pedagogical content, becomes a challenge. In our work, we propose a model to ameliorate the traditional collaborative filtering technique by (i) using the Singular Value Decomposition (SVD) to tackle the problem of scalability and data sparsity; (ii) extracting the most popular resources that the user does not interact with before to resolve the cold start problem; and (iii) combining the results of popularity and SVD factorization methods to improve the recommendation accuracy that evaluated by applying the recall, precision and f1-score of each approach. The comparison shows that the obtained results exhibit an encouraging performance of our model.
الكلمات المفتاحية :
Recommender system
E-learning
Collaborative filtering
Popularity
SVD
Y. Mediani, M. Gharzouli and C. Mediani. 2022. A Hybrid Recommender System for Pedagogical Resources. In Proceedings of the International Conference on Digital Technologies and Applications (ICDTA'22), Cham. Lecture Notes in Networks and Systems (LNNS), Saad Motahhir; Badre Bossoufi, Ed., 455, Springer, 361-371. DOI: https://doi.org/10.1007/978-3-031-02447-4_38.
APA :
Mediani, Y., Gharzouli, M. & Mediani, C. (2022). A Hybrid Recommender System for Pedagogical Resources. In Proceedings of the International Conference on Digital Technologies and Applications (ICDTA'22), Cham. In Saad Motahhir; Badre Bossoufi (Ed.). Lecture Notes in Networks and Systems (LNNS), 455, Springer, 361-371. DOI: https://doi.org/10.1007/978-3-031-02447-4_38
IEEE :
Y. Mediani, M. Gharzouli and C. Mediani, "A Hybrid Recommender System for Pedagogical Resources". In Proceedings of the International Conference on Digital Technologies and Applications (ICDTA'22), Cham. Lecture Notes in Networks and Systems (LNNS), vol. 455, Springer, pp. 361-371, 2022. DOI: https://doi.org/10.1007/978-3-031-02447-4_38.
BibTeX :
@inproceedings{misc-lab-390, author = {Mediani, Yassamina and Gharzouli, Mohamed and Mediani, Chahrazed}, title = {A Hybrid Recommender System for Pedagogical Resources}, volume = {455}, booktitle = {International Conference on Digital Technologies and Applications (ICDTA'22)}, series = {Lecture Notes in Networks and Systems (LNNS)}, location = {Cham}, isbn = {978-3-031-02447-4}, pages = {361--371}, editor = {Motahhir, Saad and Bossoufi, Badre}, publisher = {Springer}, year = {2022}, doi = {10.1007/978-3-031-02447-4\_38}, url = {https://link.springer.com/chapter/10.1007/978-3-031-02447-4\_38}, keywords = {Recommender system, E-learning, Collaborative filtering, Popularity, SVD} }
RIS :
TY - CONF TI - A Hybrid Recommender System for Pedagogical Resources AU - Y. Mediani AU - M. Gharzouli AU - C. Mediani PY - 2022 VL - 455 BT - International Conference on Digital Technologies and Applications (ICDTA'22), Cham SN - 978-3-031-02447-4 SP - 361 EP - 371 ED - S. Motahhir ED - B. Bossoufi PB - Springer AB - The present paper proposes an e-learning system that combines popularity and collaborative filtering techniques to recommend pedagogical resources. A recommender system helps users get a correct and personalized decision by applying several recommendation methods such as content-based, collaborative filtering, and other hybrid approaches. However, predicting a relevant resource with a specific context, like pedagogical content, becomes a challenge. In our work, we propose a model to ameliorate the traditional collaborative filtering technique by (i) using the Singular Value Decomposition (SVD) to tackle the problem of scalability and data sparsity; (ii) extracting the most popular resources that the user does not interact with before to resolve the cold start problem; and (iii) combining the results of popularity and SVD factorization methods to improve the recommendation accuracy that evaluated by applying the recall, precision and f1-score of each approach. The comparison shows that the obtained results exhibit an encouraging performance of our model. KW - Recommender system KW - E-learning KW - Collaborative filtering KW - Popularity KW - SVD DO - 10.1007/978-3-031-02447-4_38 UR - https://link.springer.com/chapter/10.1007/978-3-031-02447-4_38 ID - misc-lab-390 ER -