Experimento Didático De Quimiometria Para Classificação De Óleos Vegetais Comestíveis Por Espectroscopia No Infravermelho Médio Combinado Com Análise Discriminante Por Mínimos Quadrados Parciais: Um Tutorial, Parte V

Abstract

PARTIAL LEAST SQUARES DISCRIMINANT ANALYSIS: A TUTORIAL, PART V. A teaching experiment on supervised pattern recognition in chemometrics was proposed in this tutorial to introduce partial least squares discriminant analysis (PLS-DA). A new approach of the experiment published in the first tutorial of this series was revisited and employed to the classification of edible vegetable oils. The spectra of olive, canola, soybean and corn oils were obtained using an attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectrometer in the range of 600 to 4000 cm -1 . The combination of ATR-FTIR and PLS-DA classification method was able to correctly classify 100% of the validation samples. The Matlab commands, routines and functions were presented, and a didactic explanation of the concepts and interpretation of the data was provided.

ESPECTROSCOPIA DE INFRAVERMELHO PARA CARACTERIZAÇÃO GENOTÍPICA DE CLONES DE Coffea Canephora

“…Classe 1, representado pela propriedade em São Gabriel da Palha e genótipo 02; Classe 2, representado pela propriedade em Pinheiros e genótipo 02; Classe 3, representado pela propriedade em São Domingos do Norte e genótipo LB1. (SANTANA et al, 2020). Embora no estudo em questão os valores de validação cruzada apresentaram uma taxa de erro de 11%, foi possível concluir que este modelo apresentou uma considerável separação entre as classes, visto que os resultados de exatidão, taxa de acerto, especificidade, sensibilidade e acurácia ficaram todas em 100% para o teste.…”

In Silico Infrared Spectroscopy as a Benchmark for Identifying Seized Samples Suspected of Being N-Ethylpentylone

“…Thus, the answers with (i) false-positive (FP), samples of class α erroneously assigned to class β; (ii) false-negative (FN), samples of class β wrongly assigned to class α; (iii) true-negative (TN), samples of class α correctly assigned to class α; and (iv) true-positive (TP), samples of class β correctly assigned to class β were inserted [56]. With this information, the parameters' sensitivity (Equation ( 6)), specificity (Equation ( 7)), precision (Equation ( 8)), and accuracy (Equation ( 9)) were employed to assess the robustness of the created models [74][75][76].…”

Determination of adulteration in evening primrose oil capsules by ¹H NMR analysis and chemometric techniques

“…The results of NMR analyzes have the potential to generate an enormous amount of data, however, the consequence of this is the complexity of its interpretation. In this way, chemometric methods emerge as indispensable tools to assist in the interpretation of these results (HONGYU et al 2015;SANTANA et al 2020).…”