Air Quality Prediction Model of Surabaya City Using Long Short Term Memory (LSTM) Method

The city of Surabaya is one of the largest cities in Indonesia, and air quality is one of the main problems in this city. It greatly affects the quality of life, health problems, and environmental pollution. Air quality monitoring is needed to evaluate and predict air pollution concentrations that will occur accurately. In this study, time series air quality data has been obtained from the Air Quality Monitoring Device (AQM Dev) with the main air quality parameters being PM10, CO, O3, NO2, and NO. This research uses a deep learning method with the Long Short Term Memory (LSTM) algorithm, which is a development of the Recurrent Neural Network (RNN). The parameters used are hidden layer 4, hidden layer 50 neurons, batch size 5, epoch 50, Adam optimizer, and activation function using tanh. The MSE (Mean Squared Error) value for normalized data is 0.0036 and for denormalized data, the MSE value is 44.54.