Modeling the Influence of Temperature on Couscous Grain Moisture Content Using Deep Learning

In this study, a model was developed to predict the influence of temperature on the moisture content of couscous grains using a deep learning-based approach, with special attention given to artificial neural networks. Experimental data were generated by exposing couscous grains to various temperatures, including 50, 60, and 70 ℃, using the Electronic Moisture Analyzer MA 1104 machine. Several configurations were evaluated in this study, involving the manipulation of activation functions and the number of neurons in hidden layers. Each of these configurations was then subjected to rigorous evaluation using metrics such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and Coefficient of Determination (R2). Ultimately, the optimal configuration, characterized by a 2-90-60-1 architecture, was found to have excellent performance, with an R2 of 0.99988 and an MSE of 0.00374, conclusively demonstrating that the effectiveness of deep learning, particularly artificial neural networks, can be used for the complex analysis of the relationship between temperature and the moisture content of couscous grains.