Student Sentiment Analysis on the Use of Generative Artificial Intelligence Platforms to Support Academic Activities

Abstract

The development of Generative Artificial Intelligence (Generative AI) has significantly supported students' academic activities, including information retrieval, assignment completion, and learning. This study aims to analyze student sentiment toward the use of Generative AI platforms in academic contexts using the Naïve Bayes algorithm. Unlike previous studies that primarily examined social media data or focused on a single Generative AI platform, this research utilizes primary textual responses collected directly from students regarding their experiences with multiple Generative AI platforms, providing a broader understanding of student perceptions in higher education. A quantitative approach was employed using questionnaire data from 102 students. After data screening, two incomplete responses were excluded, resulting in 100 valid responses for sentiment analysis. The data underwent text preprocessing, including case folding, cleaning, tokenization, stopword removal, and stemming, followed by TF-IDF weighting before classification. The results indicate that 73% of responses expressed positive sentiment, 22% were neutral, and 5% were negative. Model evaluation achieved an accuracy of 75%, precision of 56%, recall of 75%, and an F1-score of 64%. Overall, the findings demonstrate that students generally perceive Generative AI platforms positively and consider them beneficial in supporting academic activities, highlighting their growing role in higher education learning environments.