Student textual feedback sentiment analysis using machine learning techniques to improve the quality of education

Abstract

Several educational institutions worldwide work hard to obtain student feedback to explore their views on the courses and faculty. This feedback is utilized to enhance the institution’s environment. In this modern world, institutes use data or feedback collection techniques. Still, they do not have proper techniques to analyze and use this data to improve the institute's educational quality from such textual feedback. This study presents techniques for analyzing the sentiments of student textual feedback. In this paper, machine learning methods, including Random Forest, Multinomial Nave Bayes Classifier, and Long Short-Term Memory, are applied. These methods are compared, and the experimental findings show that Long Short-Term Memory provides higher accuracy than other techniques.