| |
|
|
 | 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: |
04/07/2024 |
|
Data da última atualização: |
30/09/2024 |
|
Tipo da produção científica: |
Artigo em Periódico Indexado |
|
Autoria: |
NEIVA, D. K.; GUEDES, W. N.; MARTIN NETO, L.; VILLAS BOAS, P. R. |
|
Afiliação: |
UNIVERSIDADE DE SÃO PAULO; WESLEY NASCIMENTO GUEDES, UNIVERSIDADE ESTADUAL PAULISTA JÚLIO DE MESQUITA FILHO (UNESP); LADISLAU MARTIN NETO, CNPDIA; PAULINO RIBEIRO VILLAS BOAS, CNPDIA. |
|
Título: |
Enhancing elemental quantification in LIBS with SHAP-guided emission line analysis: A soil carbon study. |
|
Ano de publicação: |
2024 |
|
Fonte/Imprenta: |
Spectrochimica Acta Part B: Atomic Spectroscopy, v. 217, 106971, 2024. |
|
Páginas: |
1 - 8 |
|
ISSN: |
0584-8547 |
|
DOI: |
https://doi.org/10.1016/j.sab.2024.106971 |
|
Idioma: |
Inglês |
|
Conteúdo: |
ABSTRACT In laser-induced breakdown spectroscopy (LIBS), identifying key emission lines for accurate elemental quantification has long posed a challenge. Traditional methods rely on experimental knowledge, atomic databases, and intricate spectral analyses. Although machine learning techniques – such as boosting algorithms and neural networks – offer efficient processing for large datasets, the complexity of these techniques often compromises interpretability. To address this issue, our study integrates the SHapley Additive exPlanations (SHAP) algorithm with gradient boosting models in order to interpret the most important spectral features, thus enhancing our understanding of how specific emission lines contribute to the carbon (C) concentration predictions in soils. Deployed on a large dataset of 1019 soil samples, a wrapper method with a random forest regressor reduced the initial spectral intensity features from 13,748 to 1098. Subsequent application of a LightGBM regression model calibrated via the Optuna framework yielded R2 of 0.98 and 0.77, and RMSE values of 1.55 and 4.54 g kg− 1 for training and validation sets, respectively. The SHAP summary plot showed that C emission lines influenced the model's predictions positively, as anticipated, whereas silicon (Si) emission lines produced a negative impact, suggesting a lower C concentration in sandy soils. Our findings not only validate the efficacy of SHAP in improving LIBS-based soil C quantification, but they also offer a sophisticated framework for decoding the complex interplay between emission lines and target elemental concentrations. MenosABSTRACT In laser-induced breakdown spectroscopy (LIBS), identifying key emission lines for accurate elemental quantification has long posed a challenge. Traditional methods rely on experimental knowledge, atomic databases, and intricate spectral analyses. Although machine learning techniques – such as boosting algorithms and neural networks – offer efficient processing for large datasets, the complexity of these techniques often compromises interpretability. To address this issue, our study integrates the SHapley Additive exPlanations (SHAP) algorithm with gradient boosting models in order to interpret the most important spectral features, thus enhancing our understanding of how specific emission lines contribute to the carbon (C) concentration predictions in soils. Deployed on a large dataset of 1019 soil samples, a wrapper method with a random forest regressor reduced the initial spectral intensity features from 13,748 to 1098. Subsequent application of a LightGBM regression model calibrated via the Optuna framework yielded R2 of 0.98 and 0.77, and RMSE values of 1.55 and 4.54 g kg− 1 for training and validation sets, respectively. The SHAP summary plot showed that C emission lines influenced the model's predictions positively, as anticipated, whereas silicon (Si) emission lines produced a negative impact, suggesting a lower C concentration in sandy soils. Our findings not only validate the efficacy of SHAP in improving LIBS-based soil C quantification, but they also offe... Mostrar Tudo |
|
Palavras-Chave: |
Gradient boosting model; Identification of emission lines; LIBS; Machine learning; SHAP method; Soil carbon quantification. |
|
Categoria do assunto: |
-- |
|
Marc: |
null
Download
Esconder MarcMostrar Marc Completo |
|
Registro original: |
Embrapa Instrumentação (CNPDIA) |
|
|
Biblioteca |
ID |
Origem |
Tipo/Formato |
Classificação |
Cutter |
Registro |
Volume |
Status |
URL |
Voltar
|
|
|
| Registros recuperados : 1 | |
| Registros recuperados : 1 | |
|
| Expressão de busca inválida. Verifique!!! |
|
|
