<![CDATA[Objective: This study aims to analyze the application of the Long Short-Term Memory (LSTM) model in predicting demand patterns for Indonesian culinary products in online marketplaces. Method: Using monthly sales data from January 2022 to May 2024, the model was trained and evaluated with the metrics Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and R². Results: The results showed an MSE of 899.70, an RMSE of 30.00, and an R² value of 0.09, indicating that the model has limitations in capturing variations in historical data. Nevertheless, LSTM still has potential as a forecasting tool for MSME entrepreneurs in decision-making related to inventory management, production planning, and marketing strategies. Novelty: Future research is recommended to expand the dataset, incorporate external factors such as seasonal trends and promotions, and explore hybrid approaches to improve prediction accuracy.]]>
Copyrights © 2025
