Sentiment analysis or Opinion Mining is a way of solving a problem based on public opinion that is widely circulated on social media which is expressed in text form. Sentiment analysis is very helpful for the government / an agency in knowing public opinion about a policy that has just been issued without using conventional survey methods. The sentiment analysis carried out focuses on trending tweet topics on Twitter with trending topics on October 5 to 10 are #Omnibuslaw, #tolakruuciptakerja, #UUCiptaKerja, and #tolakomnibuslaw, and the trending topic on November 21 and 22 is "obl makmurkan buruh" . The sentiment analysis process is carried out after the data is obtained at the data crawling stage, followed by word cleaning in the preprocessing process, and word weighting with the TF-IDF algorithm. Sentiment analysis using the naive bayes classifier method aims to obtain a classification of public opinion on the job creation law on twitter. There are two classes in this study, there are positive and negative classes. The 2000 dataset consisting of 1400 tweets that have negative sentiments & 600 positive tweets used will be divided between training data and testing data with a ratio of 60%: 40%, 70%:30%, 80%:20%, and 90 %:10%. From the evaluation results on sentiment analysis regarding public opinion on the copyright law on Twitter, the highest accuracy value is 94% with training data used at 90%, testing data at 10%. In its implementation, the results of the sentiment test show that negative sentiment results are higher than positive sentiment.