Sentiment Analysis of Generative AI Adoption: A Comparative Study of Naive Bayes and Support Vector Machine Algorithms

The rapid adoption of Generative Artificial Intelligence (AI) has led to increasing public discussions and diverse opinions across social media platforms. This study aims to analyze public sentiment toward the adoption of Generative AI and compare the performance of the Naive Bayes and Support Vector Machine (SVM) algorithms in multi-class sentiment classification. The dataset consists of 1,125 Twitter/X posts categorized into positive, neutral, and negative sentiments. The research methodology includes text preprocessing through case folding, tokenization, stopword removal, and stemming, followed by TF-IDF feature weighting and data balancing using the Synthetic Minority Over-sampling Technique (SMOTE). Model performance was evaluated using 10-fold cross-validation to ensure reliable and robust results. The findings show that SVM achieved an accuracy of 62.13%, outperforming Naive Bayes, which achieved 52.44%, representing an improvement of 9.69%. These results demonstrate that SVM is more effective in handling high-dimensional TF-IDF feature spaces for multi-class sentiment classification and can support the development of public sentiment monitoring systems for Generative AI technologies.