Repository logo
 
Publication

Interpreting Machine Learning Models with SHAP Values: Application to Crude Protein Prediction in Tamani Grass Pastures

2025Journal article Open access
Simple item page
datacite.subject.fosCiências Agrárias::Outras Ciências Agrárias
datacite.subject.fosCiências Agrárias::Agricultura, Silvicultura e Pescas
datacite.subject.sdg02:Erradicar a Fome
datacite.subject.sdg12:Produção e Consumo Sustentáveis
datacite.subject.sdg04:Educação de Qualidade
dc.contributor.authorMonteiro, Gabriela Oliveira de Aquino
dc.contributor.authorDifante, Gelson dos Santos
dc.contributor.authorMontagner, Denise Baptaglin
dc.contributor.authorEuclides, Valéria Pacheco Batista
dc.contributor.authorCastro, Marina
dc.contributor.authorRodrigues, Jéssica Gomes
dc.contributor.authorPereira, Marislayne de Gusmão
dc.contributor.authorÍtavo, Luís Carlos Vinhas
dc.contributor.authorCampos, Jecelen Adriane
dc.contributor.authorCosta, Anderson Bessa da
dc.contributor.authorMatsubara, Edson Takashi
dc.date.accessioned2026-01-09T17:10:20Z
dc.date.available2026-01-09T17:10:20Z
dc.date.issued2025
dc.description.abstractMachine learning models such as XGBoost show strong potential for predicting pasture quality metrics like crude protein (CP) content in tamani grass (Panicum maximum). However, 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 radiation) 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 determination (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 affected 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.eng
dc.description.sponsorshipThe authors thank the Embrapa Beef Cattle, Federal University of Mato Grosso do Sul Foundation, through the Postgraduate Program in Animal Science, the National Council for Scientific and Technological Development (CNPq), the Higher Education Personnel Improvement Coordination (CAPES, Finance Code 001), and the Foundation for the Support of the Development of Education, Science and Technology of the State of Mato Grosso do Sul (FUNDECT).
dc.identifier.citationMonteiro, Gabriela Oliveira de Aquino; Difante, Gelson dos Santos; Montagner, Denise Baptaglin; Euclides, Valeria Pacheco Batista; Castro, Marina; Rodrigues, Jessica Gomes; Pereira, Marislayne de Gusmao; Itavo, Luis Carlos Vinhas; Campos, Jecelen Adriane; Da Costa, Anderson Bessa; Matsubara, Edson Takashi (2025). Interpreting Machine Learning Models with SHAP Values: Application to Crude Protein Prediction in Tamani Grass Pastures. Agronomy. ISSN 2073-4395. 15:12, p. 1-17
dc.identifier.doi10.3390/agronomy15122780
dc.identifier.issn2073-4395
dc.identifier.urihttp://hdl.handle.net/10198/35439
dc.language.isoeng
dc.peerreviewedyes
dc.publisherMDPI
dc.relation.ispartofAgronomy
dc.rights.urihttp://creativecommons.org/licenses/by-nc/4.0/
dc.subjectAlgorithms
dc.subjectSHapley additive exPlanations
dc.subjectPasture management
dc.subjectPrecision livestock farming
dc.subjectPanicum maximum
dc.titleInterpreting Machine Learning Models with SHAP Values: Application to Crude Protein Prediction in Tamani Grass Pastureseng
dc.typejournal article
dspace.entity.typePublication
oaire.citation.endPage17
oaire.citation.issue12
oaire.citation.startPage1
oaire.citation.titleAgronomy
oaire.citation.volume15
oaire.versionhttp://purl.org/coar/version/c_970fb48d4fbd8a85
person.familyNameCastro
person.givenNameMarina
person.identifier2303569
person.identifier.ciencia-id6417-7D8D-FD7E
person.identifier.orcid0000-0002-6368-8098
person.identifier.ridB-5197-2016
person.identifier.scopus-author-id56612728000
relation.isAuthorOfPublicationa7a3b08e-9d22-4faf-9224-36924d8ce7c8
relation.isAuthorOfPublication.latestForDiscoverya7a3b08e-9d22-4faf-9224-36924d8ce7c8

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Interpreting Values.pdf
Size:
1.76 MB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.75 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

CIMO - Artigos em Revistas Indexados à WoS/Scopus