ABSTRACT
Air is one of the necessities of living things. Therefore, it is necessary to have good air quality. Air pollution can cause many negative impacts on life. Therefore, it is important to know the air quality in an area. Jakarta is one of the cities with poor air quality in Indonesia and the world. During the Covid-19 pandemic, the government implemented a large-scale social restriction policy, the impact of this policy was better air quality. But now it has started to back to normal, then it is important to control air quality. There are 5 locations to measure air quality in Jakarta. The results of the independence test between the location of air quality measurements and critical variables on air pollution indicate a relationship between the two variables. Moreover, there are differences in air quality before Covid-19 and during Covid-19 based on the results of the t-test. Air quality classification was carried out in this research using machine learning methods. Because there are several levels of air quality, the classification uses a multiclass classification. Support Vector Machine (SVM) and Multi-Layer Perceptron (MLP) Classifier are used in this research. Because based on the results of literature reviews, both methods produce high accuracy. The results of this research showed a comparison of the two methods. The comparison showed that the SVM method is better than the MLP Classifier. © 2022 IEEE.