Métodos multivariados aplicados para classificação de azeite de oliva extra virgem

• Ano de defesa: 2017
• Autor(a) principal: LIMA, Iloane dos Santos
• Orientador(a): CUNHA FILHO, Moacyr
• Banca de defesa: CUNHA FILHO, Moacyr, SILVA, Ronaldo Dionísio da, OLIVEIRA, Manoel Rivelino Gomes de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• dARK ID: ark:/57462/001300000892h
• Idioma: por
• Instituição de defesa: Universidade Federal Rural de Pernambuco
• Programa de Pós-Graduação: Programa de Pós-Graduação em Biometria e Estatística Aplicada
• Departamento: Departamento de Estatística e Informática
• País: Brasil
• Palavras-chave em Português: Metabonômica; Método multivariado; Azeite de oliva
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::PROBABILIDADE E ESTATISTICA
• Link de acesso: http://www.tede2.ufrpe.br:8080/tede2/handle/tede2/7246

Metabomics is a strategy that is based on the identification of patterns of a particular biological problem, by obtaining spectroscopic / spectrometric data of a given biofluid, the use of statistics to extract this information contributes significantly to the achievement of group classification. Thus, the present work aimed at the use of the meta-monetary strategy, based on nuclear magnetic resonance spectra of hydrogen and multivariate statistical techniques of grouping (principal component analysis (PCA), Fuzzy grouping) of samples of extra virgin olive oil. Were used 40 samples of extra virgin olive oil for this study. From the spectral data matrix, we used the pre-processing normalization by summation, in the samples. From the PCA, 99.1% of the variance explained using two components only, it was not possible to observe natural clusters of the data. with the application of the Fuzzy grouping, it was verified that there was distinction of the groups in organic and common, obtaining 65% confidence. The validation made by the silhouette index, which presented s (i) of 0.73, demonstrating that the adopted grouping presents adequate strength and criteria of distinction. Thus, the fuzzy grouping method was the most indicated in the construction of a classification model of samples of extra virgin olive oil, distinguishing their different modes of production, organic and common.