##plugins.themes.bootstrap3.article.main##

Rashmi Prakashpant Bijwe

Anjali B. Raut

Abstract

With the rapid expansion of e-learning educational platforms, there is a pressing need to develop personalized course recommendation systems to keep up with student demand and improve the quality of online education. A personalized system for course recommendations built on top of content filtering techniques based on Machine Learning (ML) is introduced in this paper. This method employs the descriptive attributes of courses, such as titles, ratings, summaries, and skills, to provide personalized recommendations for users based on the similarity of course content. The characteristics from textual data are extracted using TF-IDF vectorization with cosine similarity and then ML techniques are applied to estimate course relevance. The efficacy of the proposed system is evaluated utilizing a Coursera benchmark dataset, achieving high accuracy and minimal mean square error rate in forecasting user preferences and providing substantial recommendations. The results also show that content-based filtering can improve e-learning system's user engagement and satisfaction.

##plugins.themes.bootstrap3.article.details##