Enhancing Weather Prediction Using Stacked Long Short-Term Memory Networks

Mohammad Diqi, Hamzah Hamzah, SRI HASTA MULYANI

JATISI (Jurnal Teknik Informatika dan Sistem Informasi) · 18 September 2023

In one sentence. Stacked LSTM forecasts hourly Denpasar weather (temperature, pressure, humidity, wind) from 20 years of station data, reporting average R² of 0.87 and RMSE of 0.035.

Abstract

Weather prediction is crucial in various domains, such as agriculture, transportation, and disaster management. This research investigates the Stacked Long-Short Term Memory (LSTM) for weather prediction using the Denpasar Weather Data spanning 20 years from January 1, 1990, to January 7, 2020. The dataset contains hourly weather data, including temperature, pressure, humidity, and wind speed. Our Stacked LSTM model consists of multiple LSTM layers that capture temporal dependencies and patterns in the data. Evaluating the model's performance using metrics such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and R-squared (R2), we obtain an average RMSE of 0.03471, an average MAE of 0.02718, an average MAPE of 0.05572, and an average R2 of 0.87087. These results demonstrate the effectiveness of the Stacked LSTM model in accurately predicting weather conditions. The findings have practical implications for weather forecasting applications and suggest avenues for future research, such as exploring different deep learning architectures and incorporating additional features to improve weather prediction accuracy further.

Labels

Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.