|
|
Registros recuperados : 3 | |
1. | | COSTA, A. A. da S. X.; BARROS, V. A. M. de; NORONHA, R. L. F. de; LIMA FILHO, T.; MINIM, V. P. R. Como você se sente ao tomar este café? Um estudo sobre percepções, emoções e linguagem. Cadernos de Ciência & Tecnologia, v. 41, e27350, 2024. Título em inglês: How do you feel when drinking coffee? A study on perceptions, emotions and language. Biblioteca(s): Embrapa Unidades Centrais. |
| |
3. | | NORONHA, R. L.; SOARES, M. D. R.; OLIVEIRA, I. N. de; FARHATE, C. V. V.; SOUZA, Z. M. de; OLIVEIRA, S. R. de M. Soil carbon stock predictive models on archaeological black lands - natural and transformed. In: WORLD CONGRESS OF SOIL SCIENCE, 21., 2018, Rio de Janeiro. Soil science: beyond food and fuel: abstracts. Viçosa, MG: SBCS, 2018. Não paginado. WCSS 2018. Biblioteca(s): Embrapa Agricultura Digital. |
| |
Registros recuperados : 3 | |
|
|
| Acesso ao texto completo restrito à biblioteca da Embrapa Agricultura Digital. Para informações adicionais entre em contato com cnptia.biblioteca@embrapa.br. |
Registro Completo
Biblioteca(s): |
Embrapa Agricultura Digital. |
Data corrente: |
07/12/2018 |
Data da última atualização: |
07/01/2020 |
Tipo da produção científica: |
Resumo em Anais de Congresso |
Autoria: |
NORONHA, R. L.; SOARES, M. D. R.; OLIVEIRA, I. N. de; FARHATE, C. V. V.; SOUZA, Z. M. de; OLIVEIRA, S. R. de M. |
Afiliação: |
RENATO LÓPEZ NORONHA, Unicamp; MARCELO DAYRON RODRIGUES SOARES, Unicamp; INGRID NEHMI DE OLIVEIRA, Unicamp; CAMILA VIANA VIEIRA FARHATE, Unicamp; ZIGOMAR MENEZES DE SOUZA, Unicamp; STANLEY ROBSON DE MEDEIROS OLIVEIRA, CNPTIA. |
Título: |
Soil carbon stock predictive models on archaeological black lands - natural and transformed. |
Ano de publicação: |
2018 |
Fonte/Imprenta: |
In: WORLD CONGRESS OF SOIL SCIENCE, 21., 2018, Rio de Janeiro. Soil science: beyond food and fuel: abstracts. Viçosa, MG: SBCS, 2018. |
Páginas: |
Não paginado. |
Idioma: |
Inglês |
Notas: |
WCSS 2018. |
Conteúdo: |
In the Amazon region, types of soil known as Archaeological Black Lands (ABL) present anthropic horizon A and are associated with prolonged human occupation by indigenous societies from the pre-Columbian period, where chemical and physical attributes have better quality than other types of soil in the Amazon, setting a large organic carbon reservoir. However, the conversion of these natural ecosystems into cultivated environments make emerge changes in soil carbon dynamics, often leading to a decline in soil organic carbon content. Therefore, our aim was to use data mining techniques to generate predictive models for the effect of soil use on carbon stock in natural and transformed areas of Archaeological Black Lands. We carried out our experiment in Manicoré and Apuí, Amazonas State, Brazil. After field data collection and laboratory analysis, we obtained a set of data consisting of 21 attributes, 20 predictive attributes consisting of 13 soil physical attributes, 6 soil chemical attributes, 1 soil use related attribute, and 1 response variable, referring to soil carbon stock (SCS), which is the classification target. Due to the large number of attributes, we performed a selection procedure to eliminate attributes of low correlation to the response variable. For data classification, we used the binary induction technique of the decision tree through software Weka 3.8. The results obtained showed that for the depth of 0.00-0.05 and 0.05-0.10 m, the best selected subset was determined using the Wrapper method for attribute selection. In the depth of 0.00-0.05 m, we generated a model of 79% accuracy containing only six rules, including, as the most important classification attribute was soil use. For the depth of 0.05-0.10 m, we generated an eight-rule decision tree of 74% accuracy including sand as the most important attribute. In this context, we highlight the Wrapper method efficiency to select subsets of predictive attributes, capable to generate more understandable decision trees, using a smaller number of attributes in the classification process, making it faster and with a lower computational cost. In addition, data mining techniques were efficient at providing predictive models capable to assist the decision-making process on possible management practices with the potential to conserve or increase soil carbon stock in archeological black lands. MenosIn the Amazon region, types of soil known as Archaeological Black Lands (ABL) present anthropic horizon A and are associated with prolonged human occupation by indigenous societies from the pre-Columbian period, where chemical and physical attributes have better quality than other types of soil in the Amazon, setting a large organic carbon reservoir. However, the conversion of these natural ecosystems into cultivated environments make emerge changes in soil carbon dynamics, often leading to a decline in soil organic carbon content. Therefore, our aim was to use data mining techniques to generate predictive models for the effect of soil use on carbon stock in natural and transformed areas of Archaeological Black Lands. We carried out our experiment in Manicoré and Apuí, Amazonas State, Brazil. After field data collection and laboratory analysis, we obtained a set of data consisting of 21 attributes, 20 predictive attributes consisting of 13 soil physical attributes, 6 soil chemical attributes, 1 soil use related attribute, and 1 response variable, referring to soil carbon stock (SCS), which is the classification target. Due to the large number of attributes, we performed a selection procedure to eliminate attributes of low correlation to the response variable. For data classification, we used the binary induction technique of the decision tree through software Weka 3.8. The results obtained showed that for the depth of 0.00-0.05 and 0.05-0.10 m, the best selected subset was d... Mostrar Tudo |
Palavras-Chave: |
Anthropic soils; Árvore de decisão; Data mining techniques; Decision tree; Mineração de dados; Soil management system. |
Categoria do assunto: |
X Pesquisa, Tecnologia e Engenharia |
Marc: |
LEADER 03289nam a2200265 a 4500 001 2100986 005 2020-01-07 008 2018 bl uuuu u00u1 u #d 100 1 $aNORONHA, R. L. 245 $aSoil carbon stock predictive models on archaeological black lands - natural and transformed.$h[electronic resource] 260 $aIn: WORLD CONGRESS OF SOIL SCIENCE, 21., 2018, Rio de Janeiro. Soil science: beyond food and fuel: abstracts. Viçosa, MG: SBCS$c2018 300 $aNão paginado. 500 $aWCSS 2018. 520 $aIn the Amazon region, types of soil known as Archaeological Black Lands (ABL) present anthropic horizon A and are associated with prolonged human occupation by indigenous societies from the pre-Columbian period, where chemical and physical attributes have better quality than other types of soil in the Amazon, setting a large organic carbon reservoir. However, the conversion of these natural ecosystems into cultivated environments make emerge changes in soil carbon dynamics, often leading to a decline in soil organic carbon content. Therefore, our aim was to use data mining techniques to generate predictive models for the effect of soil use on carbon stock in natural and transformed areas of Archaeological Black Lands. We carried out our experiment in Manicoré and Apuí, Amazonas State, Brazil. After field data collection and laboratory analysis, we obtained a set of data consisting of 21 attributes, 20 predictive attributes consisting of 13 soil physical attributes, 6 soil chemical attributes, 1 soil use related attribute, and 1 response variable, referring to soil carbon stock (SCS), which is the classification target. Due to the large number of attributes, we performed a selection procedure to eliminate attributes of low correlation to the response variable. For data classification, we used the binary induction technique of the decision tree through software Weka 3.8. The results obtained showed that for the depth of 0.00-0.05 and 0.05-0.10 m, the best selected subset was determined using the Wrapper method for attribute selection. In the depth of 0.00-0.05 m, we generated a model of 79% accuracy containing only six rules, including, as the most important classification attribute was soil use. For the depth of 0.05-0.10 m, we generated an eight-rule decision tree of 74% accuracy including sand as the most important attribute. In this context, we highlight the Wrapper method efficiency to select subsets of predictive attributes, capable to generate more understandable decision trees, using a smaller number of attributes in the classification process, making it faster and with a lower computational cost. In addition, data mining techniques were efficient at providing predictive models capable to assist the decision-making process on possible management practices with the potential to conserve or increase soil carbon stock in archeological black lands. 653 $aAnthropic soils 653 $aÁrvore de decisão 653 $aData mining techniques 653 $aDecision tree 653 $aMineração de dados 653 $aSoil management system 700 1 $aSOARES, M. D. R. 700 1 $aOLIVEIRA, I. N. de 700 1 $aFARHATE, C. V. V. 700 1 $aSOUZA, Z. M. de 700 1 $aOLIVEIRA, S. R. de M.
Download
Esconder MarcMostrar Marc Completo |
Registro original: |
Embrapa Agricultura Digital (CNPTIA) |
|
Biblioteca |
ID |
Origem |
Tipo/Formato |
Classificação |
Cutter |
Registro |
Volume |
Status |
Fechar
|
Nenhum registro encontrado para a expressão de busca informada. |
|
|