Sentiment Analysis of Student’s Subjective Feedback Data Using Natural Language Processing

The information processing system of the human brain is dependent not only purely upon the cognitive domain but also involves an interdependent affective or emotional domain. These interdependent domains are involved in every learning process, and the educational system is not an exception where the cognitive field deals with learning content. In contrast, the emotional or affective domain focuses on areas that give necessary mental energy. Also, according to L. Dee Fink, if students are involved in the learning process, then there is a high degree of mental energy associated with it, and the entire process has crucial outcomes or results in the form of a powerful learning experience. We can use sentiment analysis (one of the hottest topics and research domains in machine learning and Natural Language Processing) to analyze students’ affective or emotional domain by analyzing their feedback in the teaching-learning environment. Sentiment analysis is the most frequently employed technique to analyze subjective feedback in various disciplines, especially student feedback in education. It involves understanding students' feedback's meaning, tone, context, and intent by analyzing and categorizing students’ opinions into positive, negative, and neutral classes.Our study has developed a web-based sentiment analysis tool-Pratikriya using various Python-based libraries to classify the sentiments expressed by students in their respective subjective feedback data. Sentiment analysis of student feedback needs a lot of attention as a research topic because the brief personal feedback given by students contains lots of information. The earlier studies were more oriented towards binary classification, but in our research, we are oriented towards a multi-classification system for analyzing the emotional feedback of students.