Pengujian Algoritma Long Short Term Memory untuk Prediksi Kualitas Udara dan Suhu Kota Bandung
Ali Khumaidi, Ridwan Raafi’udin, Indra Permana Solihin
This study develops the LSTM modeling to predict time series data of air quality in the city of Bandung from the parameters PM10, ISPU, temperature, and humidity. Modeling LSTM with 4 hidden layers, the number of batch sizes that is 32, the optimizer is adam, value of epoch is 1000, and the loss function using Mean Squared Error. The LSTM modeling results show that the network has a fairly good performance in predicting training and testing. The modeling produces a fairly good prediction accuracy for 3 parameters (temperature, humidity, ISPU). This is indicated by the predicted RMSE value which is smaller than the standard deviation value of the dataset test. However, the prediction results generated from the four test parameters based on the order of the best are the prediction of humidity, temperature, ISPU, and PM10 with unfavorable predictive results. Penelitian ini mengembangkan…