Comparison of machine-learning integrative model and empirical equations for predicting individual dry matter intake in Girolando dairy.
No hay miniatura disponible
Fecha
Título de la revista
ISSN de la revista
Título del volumen
Editor
Resumen
Descripción
Empirical equations developed for temperate-climate conditions (NRC, 2001; NASEM, 2021) are widely used to estimate dry matter intake (DMI) of lactating cows. In tropical regions such as Brazil, diff erences in feed, genetics, and environment led to the creation of specifi c equations for Holstein and crossbred cows (BR-Leite; Oliveira et al., 2024). However, these models, built from generalized traits, have limited accuracy for predicting individual DMI. In contrast, machine-learning algorithms, such as Random Forests, have shown great potential for customized predictions by integrating multiple features at the cow level. Random Forests have been applied in several dairy studies, including the prediction of daily eating time (Foldager et al., 2020). Therefore, this study aimed to compare the DMI of Holstein × Gyr (Girolando) cows predicted by a machine-learning approach, using data- -integrated Random Forest model, with those predicted by the NRC (2001) and BR-Leite (2024).
SIMLEITE.
SIMLEITE.
Palabras clave
Raça Girolando, Modelagem, Bovino, Consumo, Ração, Matéria Seca