Читать книгу Machine Learning Algorithms and Applications - Группа авторов - Страница 25

After applying K-means clustering using Silhouette coefficient, the data is divided into seven clusters. The SVM is successfully able to classify the data into its respective air quality class with accuracy of 99%. The LSTM models for different places have been tuned accordingly to minimize MAE and RMSE. The proposed model could be used for various purposes like predicting future trends of air quality, assessing past trends of air quality, visualizing data in an effective way, issuing health advisory, and providing health effects (if any) based on the current air quality. Various parameters can be compared and it could be determined which pollutant is affecting more in a particular area and accordingly actions could be taken beforehand. Anyone could get inference from the data easily which is tough to analyze numerically and could take certain actions to control air pollution in any area.