|
|
Registro Completo |
Biblioteca(s): |
Embrapa Agricultura Digital. |
Data corrente: |
07/12/2018 |
Data da última atualização: |
28/01/2021 |
Tipo da produção científica: |
Resumo em Anais de Congresso |
Autoria: |
VAZ, G. J.; SILVA NETO, L. de F. da; OLIVEIRA, S. R. de M.; BOTELHO, F. P.; ARAUJO FILHO, J. C. de. |
Afiliação: |
GLAUBER JOSE VAZ, CNPTIA; LUIS DE FRANCA DA SILVA NETO, CNPS; STANLEY ROBSON DE MEDEIROS OLIVEIRA, CNPTIA; FABIO PEREIRA BOTELHO, CNPS; JOSE COELHO DE ARAUJO FILHO, CNPS. |
Título: |
Development of an expert system for classification of Brazilian soil profiles. |
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. Publicado também em: WORLD CONGRESS OF SOIL SCIENCE, 21., 2018, Rio de Janeiro. Soil science beyond food and fuel: proceedings. Viçosa, MG: Sociedade Brasileira de Ciência do Solo, 2019. v. I, p. 56-57. Editores: Leandro Souza da Silva, Lúcia Helena Cunha dos Anjos, Dalvan José Reinert, Heitor Cantarella, Cristine C. Muggler, Raphael B. A. Fernandes, Igor Rodrigues de Assis, Flávio A. de Oliveira Camargo. |
Conteúdo: |
The Brazilian Soil Classification System (BSCS) is the official taxonomic system for soil classification in Brazil. With the collaboration of professionals from several research and teaching institutions in the country, BSCS is in its third printed edition. Although it has been published many years ago, there is still no widely available computer program that simulates the decision-making of domain experts for the classification of Brazilian soils. The objective of this work is to build an expert system to assist professionals who need to classify Brazilian soil profiles. Based on the BSCS rules, the system simulates the reasoning of a domain expert when performing the classification of soil profiles. In addition to assisting the work of pedologists, the system can be used as a didactic resource, since it can explain in detail the path that leads to a particular solution. The system is in the prototype phase and has been developed in the Prolog language. It is able to classify soil profiles according to BSCS in different categorical levels, according to the data provided. Tests are being conducted on hundreds of samples already classified by domain experts. The development of this system brings many benefits, to wit: a) it increases the availability of knowledge on soil classification; b) it assists in the dissemination of BSCS, since it is documented not only in the form of publication, but also in software format; c) it is a rule-based system, so its development can be incremental, enabling consistency and performance tests as new knowledge is introduced; d) it has been developed using free software, resulting in reduced costs for its operation and maintenance, by the research institutions and users interested in its functionalities. Apart from these benefits, this software tool can still be used to validate previously classified profiles, classify new soil profiles and subsidize the evolution of BSCS, since it is an open taxonomic system. MenosThe Brazilian Soil Classification System (BSCS) is the official taxonomic system for soil classification in Brazil. With the collaboration of professionals from several research and teaching institutions in the country, BSCS is in its third printed edition. Although it has been published many years ago, there is still no widely available computer program that simulates the decision-making of domain experts for the classification of Brazilian soils. The objective of this work is to build an expert system to assist professionals who need to classify Brazilian soil profiles. Based on the BSCS rules, the system simulates the reasoning of a domain expert when performing the classification of soil profiles. In addition to assisting the work of pedologists, the system can be used as a didactic resource, since it can explain in detail the path that leads to a particular solution. The system is in the prototype phase and has been developed in the Prolog language. It is able to classify soil profiles according to BSCS in different categorical levels, according to the data provided. Tests are being conducted on hundreds of samples already classified by domain experts. The development of this system brings many benefits, to wit: a) it increases the availability of knowledge on soil classification; b) it assists in the dissemination of BSCS, since it is documented not only in the form of publication, but also in software format; c) it is a rule-based system, so its development can be inc... Mostrar Tudo |
Palavras-Chave: |
Brazilian Soil Classification System; Classificação de solos; Sistema especialista. |
Thesaurus Nal: |
Expert systems; Soil classification. |
Categoria do assunto: |
X Pesquisa, Tecnologia e Engenharia |
URL: |
https://ainfo.cnptia.embrapa.br/digital/bitstream/item/187974/1/Development-VAZ-WCSS-2018.pdf
https://ainfo.cnptia.embrapa.br/digital/bitstream/item/200718/1/Development-VAZ-WCSS-2019.pdf
|
Marc: |
LEADER 03228nam a2200241 a 4500 001 2100967 005 2021-01-28 008 2018 bl uuuu u00u1 u #d 100 1 $aVAZ, G. J. 245 $aDevelopment of an expert system for classification of Brazilian soil profiles.$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. Publicado também em: WORLD CONGRESS OF SOIL SCIENCE, 21., 2018, Rio de Janeiro. Soil science beyond food and fuel: proceedings. Viçosa, MG: Sociedade Brasileira de Ciência do Solo, 2019. v. I, p. 56-57. Editores: Leandro Souza da Silva, Lúcia Helena Cunha dos Anjos, Dalvan José Reinert, Heitor Cantarella, Cristine C. Muggler, Raphael B. A. Fernandes, Igor Rodrigues de Assis, Flávio A. de Oliveira Camargo. 520 $aThe Brazilian Soil Classification System (BSCS) is the official taxonomic system for soil classification in Brazil. With the collaboration of professionals from several research and teaching institutions in the country, BSCS is in its third printed edition. Although it has been published many years ago, there is still no widely available computer program that simulates the decision-making of domain experts for the classification of Brazilian soils. The objective of this work is to build an expert system to assist professionals who need to classify Brazilian soil profiles. Based on the BSCS rules, the system simulates the reasoning of a domain expert when performing the classification of soil profiles. In addition to assisting the work of pedologists, the system can be used as a didactic resource, since it can explain in detail the path that leads to a particular solution. The system is in the prototype phase and has been developed in the Prolog language. It is able to classify soil profiles according to BSCS in different categorical levels, according to the data provided. Tests are being conducted on hundreds of samples already classified by domain experts. The development of this system brings many benefits, to wit: a) it increases the availability of knowledge on soil classification; b) it assists in the dissemination of BSCS, since it is documented not only in the form of publication, but also in software format; c) it is a rule-based system, so its development can be incremental, enabling consistency and performance tests as new knowledge is introduced; d) it has been developed using free software, resulting in reduced costs for its operation and maintenance, by the research institutions and users interested in its functionalities. Apart from these benefits, this software tool can still be used to validate previously classified profiles, classify new soil profiles and subsidize the evolution of BSCS, since it is an open taxonomic system. 650 $aExpert systems 650 $aSoil classification 653 $aBrazilian Soil Classification System 653 $aClassificação de solos 653 $aSistema especialista 700 1 $aSILVA NETO, L. de F. da 700 1 $aOLIVEIRA, S. R. de M. 700 1 $aBOTELHO, F. P. 700 1 $aARAUJO FILHO, J. C. de
Download
Esconder MarcMostrar Marc Completo |
Registro original: |
Embrapa Agricultura Digital (CNPTIA) |
|
Biblioteca |
ID |
Origem |
Tipo/Formato |
Classificação |
Cutter |
Registro |
Volume |
Status |
URL |
Voltar
|
|
Registros recuperados : 5 | |
Registros recuperados : 5 | |
|
Nenhum registro encontrado para a expressão de busca informada. |
|
|