Deteksi Sentimen Publik terhadap Isu Lingkungan di Platform X (Twitter) Menggunakan Naïve Bayes dan Support Vector Machine untuk Mendukung SDGs 13: Climate Action

Abstract

Environmental issues are among the most discussed topics on social media, gaining global attention due to increasing awareness of climate change. Twitter, as one of the largest social media platforms, provides a valuable data source for understanding public perceptions of environmental topics. This study aims to detect public sentiment toward environmental issues on Twitter using the Naïve Bayes and Support Vector Machine (SVM) algorithms. The dataset used in this study is the Climate Change Twitter Sentiment Dataset from Kaggle, which contains thousands of labeled tweets related to climate change with positive, negative, and neutral sentiments. The research stages include text preprocessing (cleaning, tokenizing, and stopword removal), feature extraction using Term Frequency–Inverse Document Frequency (TF-IDF), model training, and algorithm evaluation. The experimental results show that SVM achieved higher accuracy than Naïve Bayes, with 89.4% and 84.7%, respectively. These findings indicate that SVM is more effective in identifying sentiment patterns. This research supports the achievement of Sustainable Development Goal (SDG) 13: Climate Action by leveraging artificial intelligence to analyze public opinion on environmental issues.