Success Rate Analysis of Students in MOOC Credit Courses through Supervised Learning Algorithms under Professional Discipline

The present work explicitly deals with the success rate prediction of students while opting MOOC Credit courses for various professional domains following AICTE guideline. As the four-year professional courses are subcategorized in basic science, humanities, core engineering, professional elective, and free elective; therefore, mapping of courses for success rate evaluation are performed for all the sectors for a consecutive eight batches where choice of candidates are monitored using different supervised learning algorithms. Result reflects a distinct pattern of success in nontechnical areas as well as in free and professional electives, whereas core engineering suffers a lot, owing to the major consideration of programming related courses, as evident from the results obtained through various supervised learning techniques. Comparative study is carried out based on F1 score, TP, TN, FP, FN; and regression analysis is found the best adopted technique for the same. Only SWAYAM NPTEL courses are considered for the evaluation following the guideline published in India Gazette. Artificial intelligence powered customized learning enhances the course enrollment rate as it augments re-designing and evaluation methodologies based on previous performances. Statistical findings provide improved direction about the mode of future learning, which is partially reflected in NEP 2020.