Air Quality Prediction Based on Improved LSTM Model

The quality of air has a serious impact on social and economic development and the health of the population. In order to keep track of the trend of air quality, it is especially important to forecast it accurately. In this paper, we study the application of LSTM in air quality prediction. Firstly, in order to solve the problem that LSTM only focuses on the previous information, this paper uses BiLSTM to read the before and after data simultaneously and extract the features from them when predicting air quality. Then, to address the problem that BiLSTM tends to lose information when dealing with long air quality data, this paper uses an attention mechanism to selectively use the input information by weighting the results of the hidden layer of BiLSTM. Finally, the model proposed in this paper is validated. The experimental results show that the improved LSTM model is more accurate in predicting air quality.