Please use this identifier to cite or link to this item: http://www.alice.cnptia.embrapa.br/alice/handle/doc/1182534
Full metadata record
DC FieldValueLanguage
dc.contributor.authorMONTEIRO, G. O. de A.
dc.contributor.authorDIFANTE, G. dos S.
dc.contributor.authorMONTAGNER, D. B.
dc.contributor.authorEUCLIDES, V. P. B.
dc.contributor.authorCASTRO, M.
dc.contributor.authorRODRIGUES, J. G.
dc.contributor.authorPEREIRA, M. de G.
dc.contributor.authorÍTAVO, L. C. V.
dc.contributor.authorCAMPOS, J. A.
dc.contributor.authorCOSTA, A. B. da
dc.contributor.authorMATSUBARA, E. T.
dc.date.accessioned2025-12-09T12:48:34Z-
dc.date.available2025-12-09T12:48:34Z-
dc.date.created2025-12-09
dc.date.issued2025
dc.identifier.citationAgronomy, v. 15, 2780, 2025.
dc.identifier.issn2073-4395
dc.identifier.urihttp://www.alice.cnptia.embrapa.br/alice/handle/doc/1182534-
dc.descriptionMachine learning models such as XGBoost show strong potential for predicting pasture quality metrics like crude protein (CP) content in tamani grass (Panicum maximum). How- ever, their ‘black box’ nature hinders practical adoption. To address this limitation, this study applied SHapley Additive exPlanations (SHAP) to interpret an XGBoost model and uncover how management practices (grazing interval, nitrogen fertilization, and pre- and post-grazing heights) and environmental factors (precipitation, temperature, and solar radi- ation) jointly influence CP predictions. Data were divided into 80% for training/validation and 20% for testing. Model performance was assessed with stratified 5-fold cross-validation, and hyperparameters were tuned via grid search. The XGBoost model yielded a Pearson correlation coefficient (r) of 0.78, a mean absolute error (MAE) of 1.45, and a coefficient of de- termination (R2 ) of 0.57. The results showed that precipitation in the range of 100–180 mm increased the predicted CP content. Application of 240 kg N ha−1 year−1 positively af- fected predicted CP, whereas a lower dose of 80 kg N ha−1 year−1 had a negative impact, reducing predicted levels of CP. These findings highlight the importance of integrated management strategies that combine grazing height, nitrogen fertilization, and grazing intervals to optimize crude protein levels in tamani grass pastures.
dc.language.isopor
dc.rightsopenAccess
dc.titleInterpreting machine learning models with SHAP values: application to crude protein prediction in Tamani grass pastures.
dc.typeArtigo de periódico
dc.subject.thesagroPastagem
dc.subject.thesagroPanicum Maximum
dc.subject.thesagroProteína Bruta
dc.subject.nalthesaurusCrude protein
dc.subject.nalthesaurusAlgorithms
dc.subject.nalthesaurusPasture management
riaa.ainfo.id1182534
riaa.ainfo.lastupdate2025-12-09
dc.identifier.doihttps://doi.org/10.3390/agronomy15122780
dc.contributor.institutionGABRIELA OLIVEIRA DE AQUINO MONTEIRO, UNIVERSIDADE FEDERAL DE MATO GROSSO DO SUL; GELSON DOS SANTOS DIFANTE, UNIVERSIDADE FEDERAL DE MATO GROSSO DO SUL; DENISE BAPTAGLIN MONTAGNER, CNPGC; VALERIA PACHECO BATISTA EUCLIDES, CNPGC; MARINA CASTRO, INSTITUTO POLITÉCNICO DE BRAGANÇA; JÉSSICA GOMES RODRIGUES, UNIVERSIDADE FEDERAL DE MATO GROSSO DO SUL; MARISLAYNE DE GUSMÃO PEREIRA, UNIVERSIDADE FEDERAL DE MATO GROSSO DO SUL; LUÍS CARLOS VINHAS ÍTAVO, UNIVERSIDADE FEDERAL DE MATO GROSSO DO SUL; JECELEN ADRIANE CAMPOS, UNIVERSIDADE FEDERAL DE MATO GROSSO DO SUL; ANDERSON BESSA DA COSTA, UNIVERSIDADE FEDERAL DE MATO GROSSO DO SUL; EDSON TAKASHI MATSUBARA, UNIVERSIDADE FEDERAL DE MATO GROSSO DO SUL.
Appears in Collections:Artigo em periódico indexado (CNPGC)

Files in This Item:
File Description SizeFormat 
Interpreting-machine-learning-models-2025.pdf1,79 MBAdobe PDFThumbnail
View/Open
Show simple item record

FacebookTwitterDeliciousLinkedInGoogle BookmarksMySpace