Effects of Information Technology Modularity on Content-Boosted Collaborative Filtering E-Learning Recommenders

Information technology has revolutionized e-learning and significantly impacted educational experiences by shifting them from instructor-centred to learner-centred, enabling adaptive, personalized, anytime, and anywhere learning. Modular design and technology integration underpin personalized learning models, which offer customized learning content and pathways that employ filtering techniques to enhance recommendation accuracy. Collaborative filtering in e-learning allows for personalized course offerings based on user preferences and interests, which influence student engagement and improve learning performance. The study investigates how information technology (IT) modularity influences the design and development of content-boosted collaborative filtering (CBCF) recommender systems in Ghana's higher education institutions, highlighting their importance for online learning. Using a mixed-methods approach, the experiment examines the effects of IT modularity on CBCF functionality, function, and effectiveness. The classifiers achieved a minimum accuracy of 80.6% and ROC area values ranging from 0.772 to 1.000, indicating improved classification accuracy and high-quality models. The results show that encapsulation has a moderate effect on the design and development of content-boosted collaborative filtering recommender systems, while modularity, interoperability, and separation of concerns have a high effect. IT modularity improves the adaptability, scalability, and recommendation accuracy of CBCF functions while moderately enhancing security and privacy. Recommendation accuracy and personalised learning positively influence the efficacy of CBCF recommender systems. This research suggests that modular architectures improve system functionality and accuracy by enhancing autonomous innovation, scalability, and reducing cold-start issues, despite potential computational complexity and maintenance overheads.