| |
|
|
 | Acesso ao texto completo restrito à biblioteca da Embrapa Instrumentação. Para informações adicionais entre em contato com cnpdia.biblioteca@embrapa.br. |
|
Registro Completo |
|
Biblioteca(s): |
Embrapa Instrumentação. |
|
Data corrente: |
09/06/2025 |
|
Data da última atualização: |
09/06/2025 |
|
Tipo da produção científica: |
Artigo em Periódico Indexado |
|
Autoria: |
GUEDES, W. N.; BABOS, D. V.; FREITAS, V. S.; NEIVA, D. K.; SOUZA. D.; MARTIN NETO, L.; MILORI, D. M. B. P.; VILLAS-BOAS, P. R. |
|
Afiliação: |
UNIVERSIDADE DE SÃO PAULO; LADISLAU MARTIN NETO, CNPDIA; DEBORA MARCONDES BASTOS PEREIRA, CNPDIA; PAULINO RIBEIRO VILLAS BOAS, CNPDIA. |
|
Título: |
Robust Soil Total Carbon Prediction Using LIBS: Integrating Expert Knowledge with Machine Learning and External Dataset Evaluation. |
|
Ano de publicação: |
2025 |
|
Fonte/Imprenta: |
Atomic Spectroscopy, v. 46, n. 2, 2025. |
|
Páginas: |
141–149 |
|
DOI: |
www.at-spectrosc.com/as/article/pdf/2025018 |
|
Idioma: |
Inglês |
|
Conteúdo: |
Accurate quantification of total soil carbon (C) is challenging because of the complex composition of soil samples, which include varying levels of organic matter, oxides of Al and Fe, and diverse soil textures. These factors introduce variability and spectral interference, which hinder the measurement accuracy. In laser-induced breakdown spectroscopy (LIBS), the key C emission lines often overlap with those of Al and Fe, complicating the development of reliable calibration models. This study presents a comprehensive machine learning approach for total C quantification using LIBS data. It integrates emission lines identified through the SHapley Additive exPlanations (SHAP) algorithm and literature insights with spectral preprocessing techniques such as baseline correction, peak fitting, and peak area calculations. Although the SHAP algorithm effectively identified crucial emission lines, expert knowledge was vital in refining variable selection and excluding irrelevant data. Two datasets were utilized: one containing 1019 Brazilian soil samples representing various soil types and textures, and another containing 387 samples collected later that served as an external test set. The first dataset was divided into 713 samples for model training and 306 samples for validation. The external test set was used to evaluate the reproducibility and stability of the model. The integration of expert knowledge and machine learning produced a highly efficient model that reduced the number of variables without compromising accuracy. Among the tested algorithms, the Extra Trees regression model excelled, achieving R² values of 0.74, 0.74, and 0.72 for the training, validation, and external test sets, respectively, with root mean square error values of 5.1, 5.1, and 5.0 g kg⁻¹ of total C. This approach outperformed previous machine learning models in terms of accuracy and adaptability, demonstrating its potential applicability beyond soil analysis. This study highlights a pathway for creating robust predictive models across various domains by combining machine learning, spectral processing, and expert insights. MenosAccurate quantification of total soil carbon (C) is challenging because of the complex composition of soil samples, which include varying levels of organic matter, oxides of Al and Fe, and diverse soil textures. These factors introduce variability and spectral interference, which hinder the measurement accuracy. In laser-induced breakdown spectroscopy (LIBS), the key C emission lines often overlap with those of Al and Fe, complicating the development of reliable calibration models. This study presents a comprehensive machine learning approach for total C quantification using LIBS data. It integrates emission lines identified through the SHapley Additive exPlanations (SHAP) algorithm and literature insights with spectral preprocessing techniques such as baseline correction, peak fitting, and peak area calculations. Although the SHAP algorithm effectively identified crucial emission lines, expert knowledge was vital in refining variable selection and excluding irrelevant data. Two datasets were utilized: one containing 1019 Brazilian soil samples representing various soil types and textures, and another containing 387 samples collected later that served as an external test set. The first dataset was divided into 713 samples for model training and 306 samples for validation. The external test set was used to evaluate the reproducibility and stability of the model. The integration of expert knowledge and machine learning produced a highly efficient model that reduced the number ... Mostrar Tudo |
|
Palavras-Chave: |
C sequestration; Global carbon credit; Soil carbon. |
|
Categoria do assunto: |
-- |
|
Marc: |
LEADER 02957naa a2200265 a 4500
001 2176520
005 2025-06-09
008 2025 bl uuuu u00u1 u #d
024 7 $awww.at-spectrosc.com/as/article/pdf/2025018$2DOI
100 1 $aGUEDES, W. N.
245 $aRobust Soil Total Carbon Prediction Using LIBS$bIntegrating Expert Knowledge with Machine Learning and External Dataset Evaluation.$h[electronic resource]
260 $c2025
300 $a141–149
520 $aAccurate quantification of total soil carbon (C) is challenging because of the complex composition of soil samples, which include varying levels of organic matter, oxides of Al and Fe, and diverse soil textures. These factors introduce variability and spectral interference, which hinder the measurement accuracy. In laser-induced breakdown spectroscopy (LIBS), the key C emission lines often overlap with those of Al and Fe, complicating the development of reliable calibration models. This study presents a comprehensive machine learning approach for total C quantification using LIBS data. It integrates emission lines identified through the SHapley Additive exPlanations (SHAP) algorithm and literature insights with spectral preprocessing techniques such as baseline correction, peak fitting, and peak area calculations. Although the SHAP algorithm effectively identified crucial emission lines, expert knowledge was vital in refining variable selection and excluding irrelevant data. Two datasets were utilized: one containing 1019 Brazilian soil samples representing various soil types and textures, and another containing 387 samples collected later that served as an external test set. The first dataset was divided into 713 samples for model training and 306 samples for validation. The external test set was used to evaluate the reproducibility and stability of the model. The integration of expert knowledge and machine learning produced a highly efficient model that reduced the number of variables without compromising accuracy. Among the tested algorithms, the Extra Trees regression model excelled, achieving R² values of 0.74, 0.74, and 0.72 for the training, validation, and external test sets, respectively, with root mean square error values of 5.1, 5.1, and 5.0 g kg⁻¹ of total C. This approach outperformed previous machine learning models in terms of accuracy and adaptability, demonstrating its potential applicability beyond soil analysis. This study highlights a pathway for creating robust predictive models across various domains by combining machine learning, spectral processing, and expert insights.
653 $aC sequestration
653 $aGlobal carbon credit
653 $aSoil carbon
700 1 $aBABOS, D. V.
700 1 $aFREITAS, V. S.
700 1 $aNEIVA, D. K.
700 1 $aSOUZA. D.
700 1 $aMARTIN NETO, L.
700 1 $aMILORI, D. M. B. P.
700 1 $aVILLAS-BOAS, P. R.
773 $tAtomic Spectroscopy$gv. 46, n. 2, 2025.
Download
Esconder MarcMostrar Marc Completo |
|
Registro original: |
Embrapa Instrumentação (CNPDIA) |
|
|
Biblioteca |
ID |
Origem |
Tipo/Formato |
Classificação |
Cutter |
Registro |
Volume |
Status |
URL |
Voltar
|
|
|
| Registros recuperados : 26 | |
| 2. |  | DONATO, S. L. R.; COELHO, E. F.; MARQUES, P. R. R.; ARANTES, A. DE M. Considerações ecológicas, fisiológicas e de manejo. In: FERREIRA, C. F.; SILVA, S. de O. e; AMORIM, E. P.; SEREJO, J. A. dos S. (Ed.). O agronegócio da banana. Brasília, DF: Embrapa, 2016. 47-110 il.| Tipo: Capítulo em Livro Técnico-Científico |
| Biblioteca(s): Embrapa Mandioca e Fruticultura. |
|   |
| 8. | | AZEVEDO, V. F. de; DONATO, S. L. R.; ARANTES, A. de M.; MAIA, V. M.; SILVA, S. de O. e. Avaliação de bananeiras tipo prata, de porte alto, no semiárido. Ciência e Agrotecnologia, Lavras, v. 34, n. 6, p. 1372-1380, nov./dez. 2010. Também disponível em: <http://www.scielo.br/scielo.php?script=sci_arttext&pid=S1413-70542010000600003&lng=pt&nrm=iso&tlng=pt>.| Tipo: Artigo em Periódico Indexado | Circulação/Nível: B - 1 |
| Biblioteca(s): Embrapa Mandioca e Fruticultura. |
|  |
| 9. | | ARANTES, A. de M.; DONATO, S. L. R.; SILVA, T. S.; RODRIGUES FILHO, V. A.; AMORIM, E. P. Agronomic evaluation of banana plants in three production cycles in Southwestern State of Bahia. Revista Brasileira de Fruticultura, Jaboticabal, SP, v. 39, n. 1, Mar, 2017. (e-990), Epub,| Tipo: Artigo em Periódico Indexado | Circulação/Nível: B - 1 |
| Biblioteca(s): Embrapa Mandioca e Fruticultura. |
| |
| 10. | | DONATO, S. L. R.; ARANTES, A. de M.; SILVA, S. de O. e; CORDEIRO, Z. J. M. Comportamento fitotécnico da bananeira 'Prata-Anã' e de seus híbridos. Pesquisa Agropecuária Brasileira, Brasília, DF, v. 44, n. 12, p. 1608-1615, dez. 2009 Título em inglês: Phytotechnical behavior of 'Prata-Anã' banana and progeny hybrids.| Biblioteca(s): Embrapa Unidades Centrais. |
|  |
| 13. | | MARQUES, P. R. R; DONATO, S. L. R.; PEREIRA, M. C. T.; COELHO, E. F.; ARANTES, A. de M. Características agronômicas de bananeiras tipo Prata sob diferentes sistemas de irrigação. Pesquisa Agropecuária Brasileira, Brasília, DF, v. 46, n. 8, p. 852-859, ago. 2011. Título em inglês: Agronomic characteristics of Prata type banana plants under different irrigation systems.| Tipo: Artigo em Periódico Indexado | Circulação/Nível: A - 2 |
| Biblioteca(s): Embrapa Mandioca e Fruticultura; Embrapa Unidades Centrais. |
| |
| 15. | | RAMOS, A. G. O.; DONATO, S. L. R.; ARANTES, A. de M.; COELHO FILHO, M. A.; RODRIGUES, M. G. V. Evaluation of gas exchanges and production of genotypes of maçã banana type cultivated in the semi-arid region of Bahia. Revista Brasileira de Fruticultura, Jaboticabal, v. 40, n. 3, (e-500), 2018.| Tipo: Artigo em Periódico Indexado | Circulação/Nível: B - 1 |
| Biblioteca(s): Embrapa Mandioca e Fruticultura. |
| |
| 16. | | SANTOS, M. R. dos; DONATO, S. L. R.; ARANTES, A. de M.; COELHO, E. F.; OLIVEIRA, P. M. de. Gas exchange in 'BRS Princesa' banana ( Musa spp.) under partial rootzone drying irrigation in the north of Minas Gerais, Brazil. Acta Agronómica, v. 66, n.3, p 378-384, 2017.| Tipo: Artigo em Periódico Indexado | Circulação/Nível: B - 2 |
| Biblioteca(s): Embrapa Mandioca e Fruticultura. |
| |
| 17. | | ARANTES, A. de M.; DONATO, S. L. R.; SIQUEIRA, D. L. de; COELHO, E. F.; SILVA, T. S. Gas exchange in different varieties of banana prata in semi-arid environment. Revista Brasileira de Fruticultura, Jaboticabal - SP, v. 38, n. 2, e-600, March/April 2016.| Tipo: Artigo em Periódico Indexado | Circulação/Nível: B - 1 |
| Biblioteca(s): Embrapa Mandioca e Fruticultura. |
| |
| 18. | | MARQUES, P. R. R.; DONATO, S. L. R.; SÃO JOSÉ, A. R.; ROSA, R. C. C.; ARANTES, A. de M. Nutritional status and production of Prata-Anã (AAB) and BRS Platina (AAAB) banana plants with organic fertilization. Nativa, Sinop, v. 10, n. 1, p. 60-68, 2022.| Tipo: Artigo em Periódico Indexado | Circulação/Nível: B - 2 |
| Biblioteca(s): Embrapa Agrobiologia. |
| |
| 20. | | DONATO, S. L. R.; AZEVEDO, V. F. de; ARANTES, A. de M.; MAIA, V. M.; PEREIRA, M. C. T.; SILVA, S. de O. e. Características de rendimento de bananeiras tipo prata de porte alto em condições semiáridas. In: SIMPÓSIO BRASILEIRO SOBRE BANANICULTURA, 7., 2010, Registro,SP. Atualidades e perspectivas da bananicultura sustentável. Registro: Sociedade Brasileira de Fruticultura, 2010. 1 CD-ROM. 5 p. PDF T7.| Tipo: Artigo em Anais de Congresso |
| Biblioteca(s): Embrapa Mandioca e Fruticultura. |
| |
| Registros recuperados : 26 | |
|
| Nenhum registro encontrado para a expressão de busca informada. |
|
|
