In 2019, 73.7% of Indonesians said that their purpose for using the internet was to access social media. One of the social media used is Twitter. With many tweets that have been published through Twitter, these tweets can contain user opinions on a particular thing, it can be like an event in the surrounding Indihome. Through Twitter, users can discuss their complaints or satisfaction with the Indihome service. For that reason, a method is needed, namely sentiment analysis to find out whether the data contains negative or positive opinions. The author uses the Naïve Bayes method in conducting sentiment analysis on the opinions or opinions of Indihome service users on Twitter, to know how accurate the Naïve Bayes method is applied to sentiment analysis. After testing using the Naïve Bayes method, the results obtained are 82% accuracy, 78% precision, 84% recall, and 81% f1-score.
Copyrights © 2022