Estimation of Air Quality Index using AI and ML Techniques

The air pollution has caused a critical threat to the environment and human health, making accurate and timely air quality monitoring very imperative. This research work is presenting an innovative approach of estimation of Air Quality Index (AQI) based on the Artificial Intelligence and Machine Learning techniques implemented in MATLAB. The AQI is a metric which quantifies the air pollution level to guide public health policies. The research begins by gathering the secondary air quality data from available sources. This data covers various parameters viz. SO
2
, NO
2
, CO, O
3
, PM10, and PM2. 5. These datasets have been preprocessed to cope up with outliers and missing values, guaranteeing the reliability of the analysis to be done. The decision tree is used in MATLAB as a method to establish a non-linear prediction model for AQI estimation. Additionally, a GRNN model is implemented in MATLAB as an alternative way of AQI estimation. This technique is based on artificial neural networks, and has proved to be accomplishing a variety of regression tasks successfully. For evaluating the performance of developed models, the testing is conducted using RMSE (Root Mean Square Error) metric. The comparative analysis presented reveals the accuracy of each model.