<![CDATA[This research explores the application of Machine Learning using Multiple Linear Regression (MLR) and Support Vector Machine (SVM) algorithms to predict air quality categories in Jakarta based on key pollutant parameters, such as PM10, PM2.5, NO2, CO, SO2, and O3. The dataset used comes from ISPU data measured from five Air quality monitoring stations in DKI Jakarta Province in 2021. The research process includes data collection, data cleaning, model implementation using the scikit-learn library, and model performance evaluation using Accuracy, R-Squared, RMSE, and MAE metrics. The results of model performance evaluation show that SVM performs better than MLR, as evidenced by higher accuracy value (91.78% vs. 90.41%), higher R-squared value (69.63% vs. 64.56%), lower RMSE value (0.2867 vs. 0.3097), and lower MAE value (0.0822 vs. 0.0959), indicating that the error in SVM model is smaller than MLR. This study proves the effectiveness of machine learning-based models in providing accurate air quality category predictions, although there are still challenges in predicting the “Good” category that require further development, such as balancing data and advanced feature engineering to improve the prediction accuracy of all categories.]]>
Copyrights © 2025
