Air pollution is the contamination of indoor or outdoor environments with chemical, physical, or biological substances that change the natural properties of the atmosphere. Domestic incinerators, cars, motorbikes, combustion products from factory processing, waste burning, and forest fires are common sources of air pollution. In Indonesia, there is no doubt that air pollution occurs because of the many forest fires in Indonesia. As a result of this case, many people's opinions differ. Various sentiments occur in cyberspace, one of which is Twitter. Twitter is the social media that accommodates the most various kinds of positive, negative and neutral opinions. Therefore, researchers want to solve the problem by implementing the SVM and Random Forest algorithms. The dataset was obtained from scrapping results using tweet harvest. The data obtained was 5545 tweets. By dividing the dataset model by 80% and 20%, the results showed that the accuracy of the SVM algorithm was better than the Random Forest algorithm. The accuracy of the SVM algorithm is 83% while the Random Forest algorithm is 81%.