Course Recommendation System Using Machin Learning

This is the authors' abstract. We don't add key points for this paper.

Today, many online educational platforms offer a large number of courses, which creates confusion for students while choosing the right course. To solve this problem, this paper proposes a simple and effective hybrid Course Recommendation System using Machine Learning and Natural Language Processing. The system combines two main techniques: Content-Based Filtering and Popularity-Based Filtering. In Content-Based Filtering, tools like TF-IDF, Count Vectorizer and Cosine Similarity are used to match course details with student interests. In Popularity-Based Filtering, courses are selected based on top subscriber count and positive user reviews. The backend is built using Python Flask with a clear user interface for easy navigation. The system was tested with multiple machine learning algorithms, where the Decision Tree model achieved the highest accuracy of 90%. Overall, this system helps students quickly find useful courses without wasting time.