Air Quality Analysis Using Random Forest Algorithm and SMOTE Data Balancing Technique

Air quality is an important element that has a direct impact on the health and well-being of living things. Polluted air, especially in urban areas, can trigger various serious health problems. Jakarta, as one of the largest cities in Indonesia, faces a major challenge in maintaining healthy air quality for its citizens. Therefore, air quality analysis in this city is very important to combine existing pollution levels. This study was conducted through analysis of Jakarta air quality data from early 2010 to the end of 2022, with the aim of classifying pollution levels based on several pollutants. Data analysis uses Random Forest, which is an ensemble algorithm of decision trees that has proven effective in handling large-dimensional data. In addition, to overcome the data imbalance contained in the air quality category, the Synthetic Minority Over-sampling Technique (SMOTE) technique is applied which functions to balance data distribution. The results of this study show that the Random Forest method is able to provide high accuracy of 99.3%. This finding shows that Random Forest is very effective in classifying air quality in Jakarta, while emphasizing the importance of the data balancing stage with the SMOTE technique in handling data smoothness. Random Forest and SMOTE are the right combination and have been proven to provide optimal results, so it is hoped that this can be input for the Jakarta regional government in dealing with air pollution.