Improving the computational efficiency of the successive projections algorithm by using a sequential regression implementation: a case study involving nir spectrometric analysis of wheat samples

Autor(es)

Soares, Anderson S. ^[1] ; Galvao Filho, Arlindo R. ^[1] ; Galvao, Roberto K. H. ^[1] ; Araujo, Mario Cesar U. ^[2]

Afiliação do(s) autor(es)

^[1] Inst Tecnol Aeronaut, Div Engn Elet, BR-12228900 Sao Jose Dos Campos, SP - Brazil

^[2] Univ Fed Paraiba, CCEN, Dept Quim, BR-58051970 Joao Pessoa, Paraiba - Brazil

• Tipo de documento: Artigo Científico
• Fonte: Journal of the Brazilian Chemical Society; v. 21, n. 4, p. 760-763, 2010.
• Citações Web of Science: 5

Resumo

This short report proposes a sequential regression implementation for the successive projections algorithm (SPA), which is a variable selection technique for multiple linear regression. An example involving the near-infrared determination of protein in wheat is presented for illustration. The resulting model predictions exhibited a correlation coefficient of 0.989 and an RMSEP (root-mean-square error of prediction) value of 0.2% m/m in the range 10.2-16.2% m/m. The proposed implementation provided computational gains of up to five-fold. (AU)

• Processo FAPESP: 07/57803-7 - Deteccao e diagnostico de falhas empregando tecnicas de classificacao de padroes com selecao de atributos.
• Beneficiário: Anderson da Silva Soares
• Modalidade de apoio: Bolsas no Brasil - Doutorado
• Processo FAPESP: 06/58850-6 - Diagnóstico, prognóstico e acomodação de falhas em sistemas dinâmicos
• Beneficiário: Takashi Yoneyama
• Modalidade de apoio: Auxílio à Pesquisa - Temático