MOOCRec_Sys: An Empirical E-learning Paradigm on Online Open Course using Sentiment Analysis and OWA operator

Abstract COVID Pandemic has brought radical transformations in the personal lives of people, like economic, social, cultural, educational. Among these, e-learning educational transformation is a very crucial paradigm of a learner’s life. This strong inclination of paradigm shift to online education has brought novel changes to the betterment of the education system. In Massive Open Online Courses (MOOC) are capable of giving promising avenues to e-teaching. Among the challenges for the learners in MOOC, one is to correctly choose the best learning option for a particular Course. In Technology Enhanced Learning, another concern is the appropriate searching of a leaning resource. This gives rise to the use and deployment of a recommender system. In this paper, we propose and evaluate a two-stage approach where first, online user reviews are considered for the course evaluation and second, User-Item Matrix is built for finding similar users. The proposed model suggests the best suitable resource to the learner based on his/her personal ability. In MOOC_RecSys, Naïve Bayes classification is used with multinomial Classifier to give a rating to reviews. The accuracy of the proposed method is encouraging and the computed RMSE for the recommendation system in 0.76.