Predictive Analysis and Performance Evaluation Using in Deep Learningbased Air Quality Monitoring System

Abstract

Air quality monitoring is vital for safeguarding the environment and public health. It plays a critical role in environmental conservation and public health management. This research explores the potential of deep learning models to enhance air quality prediction accuracy and offer valuable insights into pollution trends. We explore various deep learning methods, including Convolution Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, Auto-encoders, Generative Adversarial Networks (GANs), and Transformer-based models on Delhi Air Quality Data Set. To evaluate the models' performance based on standard metrics such as Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and accuracy. We find that Convolution Neural Networks excel in predictive accuracy and image-based air quality assessment.