<![CDATA[The increase in the price of pertalite fuel in Indonesia has caused various reactions from the public, which are widely expressed through social media such as twitter. Fuel oil (BBM) is one of the basic needs that is very important for the Indonesian people because it plays a major role in supporting various daily activities. BBM is used not only as fuel for motor vehicles, but also as a source of energy for various industrial equipment. This study aims to classify public sentiment towards the issue using the naive bayes algorithm and Gradient Boosting Machine (GBM) as a classification method. The data used in this study were obtained from the Kaggle platform, which contains a collection of tweets related to the issue of fuel prices, especially pertalite. The analysis process begins with text preprocessing, such as data cleaning, tokenization, stopword removal, and stemming. The data already has a sentiment label (positive, negative, neutral) and is divided for model training and testing. The evaluation results show that the GBM algorithm is able to classify sentiment with an accuracy rate of 60% while the Naive Bayes algorithm has an accuracy rate of 90%. These results prove that naive bayes has a higher level of accuracy than the GBM algorithm, so it can be used in processing text data from social media to understand public opinion on government policies, especially regarding fuel price increases.]]>
Copyrights © 2025
