<![CDATA[Forecasting models with high accuracy become more important during uncertain conditions, such as climate change, that could have a high effect. The forecast model's accuracy in predicting cocoa crop yield must be high to determine decision-making in management. Seven different potential predictor variables have been analyzed in this research to see the influence of cocoa crop yield. Using a scatter plot diagram, six of seven variables, relative humidity, maximum temperature, minimum temperature, evapotranspiration, rainfall, and soil moisture, are proven to influence cocoa crop yield. Then, those datasets are divided into training and validation sets using multiple linear regression analysis and a Long Short-Term Memory (LSTM) network. The output model of those methods is assessed using two metrics: coefficient of determination and Root Means Square Error (RMSE). From those model performance metrics, LSTM outperformed multiple linear regression analysis. LSTM has an R-square of 98% and an RMSE of 0.3 while multiple linear regression just reached 82% of the R-square and 2.57 of the RMSE. The LSTM model has been proven to be valid.]]>
Copyrights © 2024
