Evaluation of Diversity and Novelty of Recommendation System on Digital Learning Platform Based on User Usability and Satisfaction

There are many digital learning platforms (e-learning) today, such as Duolingo and Udemy, that offer a large amount of content, making it difficult for users to find materials that suit their learning needs. Due to this, recommendation systems were implemented to improve user satisfaction and usability. Although traditional recommender systems focus on accuracy. Other important aspects of the recommendation system used in this research highlight the importance of diversity and novelty by offering diverse and novel recommendations that introduce users to unfamiliar or surprising recommendations based on user preferences and existing content on the platform. Using k- means clustering, Intra-List Diversity (ILD), and Mean Self-Information (MSI) metrics, this research evaluates the effectiveness of the system. The results show an ILD of 0.616044328 and an MSI of 0.628135613, which indicates diverse and novel recommendations. User feedback on the user satisfaction questionnaire survey further confirmed satisfaction and usability in highlighting the value of integrating diversity and novelty into the recommendation system on the e-learning platform.