Learning-Related Sentiment Detection, Classification, and Application for a Quality Education Using Artificial Intelligence Techniques

Quality education is one of the primary objectives of any nation -build-ing strategy and is one of the seventeen Sustainable Development Goals (SDGs) by the United Nations. To provide quality education, delivering top-quality con-tent is not enough. However, understanding the learners' emotions during the learning process is equally important. However, most of this research work uses general data accessed from Twitter or other publicly available databases. These databases are generally not an ideal representation of the actual learning process and the learners' sentiments about the learning process. This research has col-lected real data from the learners, mainly undergraduate university students of dif-ferent regions and cultures. By analyzing the emotions of the students, appropriate steps can be suggested to improve the quality of education they receive. In order to understand the learning emotions, the XLNet technique is used. It investigated the transfer learning method to adopt an efficient model for learners' sentiment detection and classification based on real data. An experiment on the collected data shows that the proposed approach outperforms aspect enhanced sentiment analysis and topic sentiment analysis in the online learning community.