Predicting Air Quality: Leveraging Machine Learning for Accurate Pollution Forecasting

Predicting Air Quality: Leveraging Machine Learning for Accurate Pollution Forecasting

G. Hampika

B.Tech, professor, Dept. of ECE Institute of Aeronautical Engineering Hyderabad, India hampika.g@iare.ac.in

0000-0002-4426-998X

Bhavana Penumetcha

B.Tech, student, Dept. of CSE Institute of Aeronautical Engineering Hyderabad, India 21951a0525@iare.ac.in

0009-0007-4252-3810

D. Sai Praneeth Reddy

B.Tech, student, Dept. of CSE Institute of Aeronautical Engineering Hyderabad, India 21951a05h2@iare.ac.in

0009-0001-0142-8346

Meghana Yadavalli

B.Tech, student, Dept. of CSE Institute of Aeronautical Engineering Hyderabad, India 21951a05a0@iare.ac.in

0009-0004-7400-4175

Abstract— Air quality index (AQI) calculations must be quick and precise since air pollution poses a hazard to both the environment and public health. This work introduces a neural network model-based deep learning approach for AQI estimation. The study makes use of daily measurements of a range of pollutants from different cities, including PM2.5, PM10, NO, NO2, CO, and others. The data is put through pre-processing procedures including modelling and missing value management in order to improve the model's efficacy. A neural network consisting of two hidden layers is trained using the data processing. Using mean squared error (MSE), the model was assessed after more than 150 training runs. Based on pollution data, the research findings demonstrate that the model can forecast AQI.

Keywords—Air Quality Index (AQI), Neural Network, Machine Learning, Predictive Modeling, Mean Squared Error (MSE).