Improve the Accuracy of the System SENTIMENT ANALYSIS FOR Students about Teaching Using (GWO-CNN)

Abstract

    Sentiment analysis has become a vital area of research in artificial intelligence, particularly within natural language processing. With the widespread sharing of opinions and sentiments online, organizations, businesses, and governments can leverage automated tools to analyze feedback and evaluations. In the context of education, the increasing focus on student engagement and attendance has highlighted the importance of understanding feedback in institutional settings. This study employs a lexical sentiment analysis approach to determine the polarity of student responses, using the Vietnamese Student Feedback Corpus (UIT-VSFC), which contains 16,175 sentences of student feedback. The dataset was translated into English for analysis. The proposed method achieved an impressive accuracy of 98%. Additionally, machine learning techniques, including Naïve Bayes (NB), K-Nearest Neighbors (KNN), and Support Vector Machines (SVM), were applied, resulting in accuracies of 94%, 88%, and 92%, respectively.