Students’ Sentiment and Feedback Analysis on Online Learning System during COVID-19 Pandemic

Abstract

After the COVID-19 pandemic, no one refutes the importance of online learning in the educational process. Measuring student engagement is a crucial step towards online learning. Sentiment analysis has been widely applied in many domains, including business, social networks, and education. Particularly in the education domain, where dealing with and processing students’ opinions is a complicated task due to the nature of the language used by students and the large volume of information, sentiment analysis is growing yet remains challenging. An online learning system can automatically adapt to learners’ emotions and provide feedback about their motivations. The researchers’ challenge is how to know the student behaviour by feedback text. Most existing techniques of sentiment analysis focus only on the abstract level, broadly classifying sentiments into positive, neutral, or negative, and lacking the ability to perform fine-grained sentiment analysis. This study proposes a supervised ML-based sentiment analysis model that includes two main machine learning models. The first one is sentiment detection. There is two distinct (0 and 1) label as positive and negative. And the second one is emotional detection. There are six mains (1-6) labelled as Anger, Joy, Surprise, Disgust, Sadness, Fear. Most of the current ML-based Sentiment Analysis models used Support Vector Machine, Naive Bayes algorithms, Logistic Regression and experiment with other algorithms such as Decision Tree and Random Forest also