<![CDATA[Climate change, marked by increasing uncertainty in weather patterns, poses significant challenges to agricultural productivity in Indonesia. Accurate weather forecasting is therefore essential to support sustainable agrarian efficiency, parti-cularly in determining planting schedules, irrigation management, and mitigating crop failure risk. This study aims to apply artificial intelligence (AI) algorithms to improve weather prediction accuracy by comparing the performance of Long Short-Term Memory (LSTM), Random Forest (RF), and conventional methods, including ARIMA and Least Squares Method (LSM). The dataset used consists of historical meteorological parameters, including temperature, humidity, rainfall, and wind speed. The research process involved data collection, preprocessing, model development, and evaluation using RMSE, MAE, and R². The results reveal that LSTM outperforms the other models with an R² of 0.92, RMSE of 0.15, and MAE of 0.11, while RF achieved an R² of 0.85. In contrast, ARIMA and LSM showed lower performance with R² values below 0.80, highlighting the limitations of conventional approaches in capturing non-linear weather patterns. These findings confirm that the application of AI algorithms, particularly LSTM, provides more accurate weather forecasting, which directly contributes to sustainable agricultural practices by improving efficiency and resilience to climate variability.]]>
Copyrights © 2026
