Application of Collaborative Filtering and Explainable AI Methods in Recommendation System Modeling to Predict MOOC Course Preferences.

Praseptiawan, Mugi and Putri, Nabila Muthia and Muchtarom, M. Fikri Damar and Zakaria, Mohd Hafiz and Che Pee @ Che Hanapi, Ahmad Naim (2024) Application of Collaborative Filtering and Explainable AI Methods in Recommendation System Modeling to Predict MOOC Course Preferences. In: 2nd International Symposium on Information Technology and Digital Innovation: Creative Trends in Sustainable Information Technology Design and Innovation, ISITDI 2024.

Text
Application of Collaborative Filtering and Explainable AI Methods in Recommendation System Modeling to Predict MOOC Course Preferences.pdf
Download (515kB)

Official URL: https://10.1109/ISITDI62380.2024.10797073

Abstract

Abstract—Unquestionably, the creation of MOOCs has transformed education by making learning accessible and inexpensive for people all over the world. However, consumers frequently find it difficult to select a course that fits their needs and interests due to the vast curriculum that MOOCs offer. This paper describes how a recommendation system that applies the Collaborative Filtering approach can be used to address these kinds of problems. This paper proposes a recommendation system modeling for MOOC platforms that allow users to be recommended courses based on their preferences. Using user interactions and choices, the Collaborative Filtering approach suggests courses that are specifically tailored to everyone. Utilizing a user-based method called collaborative filtering, users' preferences are predicted by comparing them to other users. The recommendation system's transparency and interpretability are improved by the incorporation of XAI using LIME. By gaining understanding of the reasoning behind course recommendations, users can build trust and make well- informed decisions. A quantitative assessment of the prediction accuracy is provided by the recommendation system's performance evaluation utilizing the RMSE measure. Over time, this statistic aids in system improvement and raises the caliber of course recommendations.

Item Type:	Conference or Workshop Item (Paper)
Divisions:	Faculty of Information and Communication Technology
Depositing User:	NURHASHIRAH BORHAN
Date Deposited:	31 Jul 2026 07:57
Last Modified:	31 Jul 2026 07:57
URI:	http://eprints.utem.edu.my/id/eprint/30143
Statistic Details:	View Download Statistic

Actions (login required)

View Item