Deep Learning Approach to Estimate the Maize Yield Prediction Using Data From Cameroon
Abstract
Corn cultivation plays a crucial role in the Cameroon's food production, providing an important source of food and income for many farmers. However, climatic variability and unstable agricultural conditions can have a significant impact on the yield of agricultural products in general and that of maize in particular; accurately predicting these yields in different regions of Cameroon remains a difficult process due to the uncertain evolution of climatic data.. This is where deep learning comes in, a powerful approach to analyzing large amounts of data and generating predictive models. This study aims to estimate maize yield forecasts in Cameroon using geospatial climate data parameters such as temperature, precipitation, wind speed and sun exposure as well as agricultural data. The study results, based on performance evaluations of o GRU model, have 24751 in 200 epochs for GRU, a mean absolute percentage error (MAPE) of 237%, and a root mean square error (RMSE) of 518 for GRU, which demonstrates the effectiveness of the deep learning approach in predicting corn yield.