Sentiment Analysis of Bamboo Charcoal: Comparing Machine Learning Algorithms for Effective Insights

This research aims to analyze sentiments toward bamboo charcoal on social media, with a focus on public perception in the global market in English. Using data collected from the social media platform X, this study applies various machine learning algorithms, including Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Deep Learning, Naïve Bayes, Decision Tree, and Gradient Boosted Trees, with TF-IDF as the text representation. The analysis reveals that the SVM model achieved the most accurate result of 92.33%, demonstrating its effectiveness in sentiment detection. The study also found that the KNN model performed well, achieving an accuracy of 92.26%, although slightly lower than SVM. These findings highlight the growing interest in bamboo charcoal as a sustainable product, reflecting positive sentiments in the data. Additionally, the Deep Learning model also showed promising results, although it was slightly less effective than SVM and KNN. However, there were also notable concerns regarding the environmental impact of bamboo harvesting, which were primarily expressed in posts. The Decision Tree model, while useful, did not perform as well as the other models, indicating the need for further refinement. Future research could explore a broader range of social media platforms, models, and languages to gain a more comprehensive understanding of global perceptions. Furthermore, integrating sentiment analysis with real-time monitoring could help stakeholders respond more effectively to shifts in public opinion.