Fuel oil is needed as a support in life. Local fuel must be adjusted to international fuel prices so that the country's fiscal sustainability remains safe and not threatened. This price adjustment is carried out by the government as an effort to optimize the use and supply of fuel and to overcome the occurrence of a fuel crisis in the future. On the Twitter platform, the discussion about the fuel price increase even has become a trending topic due to the number of tweets discussing the issue. The number of opinions about the fuel price increase makes it difficult to determine the sentiment of the tweet manually. Therefore, sentiment analysis is needed that can classify the tweet whether it tends to be positive or negative. In this case, this analysis is mediated by the Naïve Bayes algorithm to classify the problem. Based on the sentiment analysis made, it can be seen that the Naïve Bayes method or algorithm can analyze tweets with good results. The accuracy generated in this sentiment analysis is 85% with a division of 80% training data and 20% test data. With the acquisition of these accuracy results, it can be said that the proposed algorithm has a fairly good diagnostic level. Keywords: sentiment analysis, Twitter, fuel oil, Naïve Bayes
Copyrights © 2023