Published Paper Details:

With the rapid increase in the popularity of learning platforms on the Internet, there are huge amounts of educational content available. It can be a big problem for users who wish to choose proper courses based on their own needs. There are a variety of platforms like Coursera, Udemy, or edX which provide a large amount of content related to different courses. The keyword-based searching approach used by these platforms does not provide enough accuracy for the recommendation of courses that fit perfectly into users' goals, skills, or interests. Therefore, this paper describes an intelligent course recommender system using a machine learning algorithm, which will help learners discover appropriate courses much more efficiently. In this research, a content-based filtering approach is adopted to recommend courses. First, course descriptions are transformed into numeric vectors using Term Frequency-Inverse Document Frequency (TF-IDF). Then, cosine similarity is applied to measure the similarities between courses. Such a system can recognize courses that are similar in topic to each other. The application is implemented in Python using the Flask framework. Evaluation results show that the suggested recommender system indeed works better in discovering new courses than existing approaches.