Classification of Youtube User sentiment on 5G Technology Videos with Naïve Bayes
Taupik Hidayat, Agus Bahtiar, Kaslani · Journal of Artificial Intelligence and Engineering Applications (JAIEA) · 2025
The rapid development of 5G technology has triggered various reactions from the public on social media platforms such as YouTube. User sentiment towards videos discussing 5G technology varies, from positive to negative. This research aims to improve the sentiment classification model of YouTube user reviews of videos about 5G technology with the Naïve Bayes algorithm, which is known to be able to handle large text data and sentiment variations. This research goes through the stages of collecting review data from YouTube, data preprocessing including tokenization, stop word removal, and stemming, and sentiment classification into positive, neutral, and negative categories using Naïve Bayes. The bag-of-words (BOW) technique is used to improve the algorithm's performance. The results showed a sentiment distribution of 1,581 neutral, 1,165 positive, and 517 negative. The proposed model achieved 98% accuracy, with macro average precision 0.99, recall 0.98, and F1-score 0.98. Weighted average resulted in precision 0.98, recall 0.98, and F1-score 0.98. These results show the model performs very well in sentiment classification. This research is expected to make a significant contribution in understanding public perception of 5G technology.