<![CDATA[In Indonesia itself, fuel is a very important raw material for society, especially for the industrial sector. The fuel price hike policy sparked controversy on social media, one of which was Twitter. After the increase in fuel prices was passed, every day on Twitter was filled with tweets with the hashtag (#bbmnaik). The pros and cons that exist in the community regarding the increase in fuel prices is an interesting research material. This study aims to analyze public sentiment whether it is negative or supportive. The method used is Logistic Regression assisted by the Confusion Matrix for evaluation calculations. The advantage of this method compared to other methods is that the Logistic Regression method is often used to create a predictive model whose result values are in the form of yes/no, true/false, thus this method is very suitable for this research. The data used is 3000 data with keywords (increase in fuel prices). The results of the analysis that has been carried out show that positive sentiments get an accuracy value of 38% and negative sentiments of 80%. Classification performance of the Logistic Regression method gains 73%. The results of evaluation calculations with the Confusion Matrix using data testing as many as 600 data get an accuracy rate of 77%, a precision value of 95%, a recall value of 79%, and an f1 score of 86%. So it can be concluded from the results of the sentiment analysis that has been done that the public is more pro against the rejection of the increase in fuel prices.]]>
Copyrights © 2023
