WEATHER FORECAST FROM TIME SERIES DATA USING LSTM ALGORITHM

Accurate weather forecasts play an important role in today's world as various sectors such marine, navigation, agriculture and industry are basically dependent on conditions. Weather also used to predict the occurrence of natural disasters. forecasting determines exact value parameters then predicts future In this study are. Different were collected from Serang Maritime Meteorological Station analyzed using a neural network-based algorithm, namely Long-short term memory (LSTM). predicting conditions LSTM networks trained combination different parameters, temperature, humidity, rainfall, wind speed. After training model these predictions performed. The prediction results evaluated RMSE. Prediction show that is more accurate when temperature data with RMSE 0.37, speed 0.72, sunlight 2.79, humidity 5.05. This means very good at studying data, inversely proportional data.