Enhancing Collaborative Filtering Recommendations for Web-based Learning Platforms with Genetic Algorithms

Web-based learning platforms are now offering a considerable amount of training options, exhibiting great variability, similar to the one encountered by a potential customer in electronic shops. To mitigate this, efficient and effective recommendations engines are needed, capable of satisfying the specific needs of each user, while achieving the best possible promotion of available online training. This paper discusses the design of a potential generic architecture for online education recommender systems, specifically targeted for promoting online courses and web-based learning material. From an algorithmic perspective, the system relies on item-based and user-based collaborative filtering approaches. It extends this approach with a genetic algorithm, thus increasing its potential impact. Overall, the paper paves the ground for the specification of generic principles governing the design of personalized online education platforms as well as identifying metrics for evaluating their performance.