Comparison of machine learning and deep learning techniques for the prediction of air pollution: a case study from China

Abstract The adverse effect of air pollution has always been a problem for human health. presence high level pollutants can cause severe illnesses such as emphysema, chronic obstructive pulmonary disease (COPD), or asthma. Air quality prediction helps us to undertake practical action plans controlling pollution. Quality Index (AQI) reflects the degree concentration in locality. average AQI was calculated various cities China understand annual trends. Furthermore, index predicted ten major across using five different deep learning techniques, namely, Recurrent Neural Network (RNN), Bidirectional Gated unit (Bi-GRU), Long Short-Term Memory (BiLSTM), Convolutional BiLSTM (CNN-BiLSTM), and (Conv1D-BiLSTM). performance these models compared with machine model, eXtreme Gradient Boosting (XGBoost) discover most efficient model. results suggest that XGBoost, outperforms models. While Conv1D-BiLSTM CNN-BiLSTM perform well among estimation (AQI), RNN Bi-GRU are least performing ones. Thus, both XGBoost neural network capable capturing non-linearity present dataset reliable accuracy.