Opinion Mining from Student Feedback Data Using Supervised Learning Algorithms

This paper examines the mining of ideas using learning algorithms monitored to determine student polarity feedback based on previously defined aspects of teaching, learning, course content, examination pattern, laboratory, library facilities, and extra co-curricular activities. The research involves using an aggregate of Machine Learning (ML) and Natural Language Processing (NLP) strategies with open student reaction information gathered from the University's class module survey results. Furthermore, this paper imparts a look at the overall performance of Naive Bayes algorithms. The survey data are used for processing facts and to train algorithms for binomial separation. Trained algorithms are also able to predict the polarity of a scholar's opinion, based totally on excluded features which include coaching, mastering, direction content, examination sample, laboratory, library centres, more co-curricular sports, and many others.