Two-Step Partial Least Squares-Discriminant Analysis Modeling for Accurate Classification of Edible Sea Salt Products Using Laser-Induced Breakdown Spectroscopy

Abstract

Laser-induced breakdown spectroscopy (LIBS) has been widely applied to material classification in various fields, and partial least squares-discriminant analysis (PLS-DA) is one of the frequently used classical multivariate statistics to construct classification models based on the LIBS spectra. However, classification accuracy of the PLS-DA model is sensitive to the number of classes and their similarities. Considering this characteristic of PLS-DA, we suggest a two-step PLS-DA modeling approach to improve the classification accuracy. This strategy was demonstrated for a 6-class problem in which six commercial edible sea salts produced in Japan, South Korea, and France are classified using their LIBS spectra. At the first step, test spectra were sorted into four classes and one extended class, composed of the two other most confusing classes, and then the test spectra in the extended class were further classified into each of the two constituent classes which were modeled separately from the other four classes. This two-step classification has been found to remarkably improve the PLS-DA classification accuracy by maximizing the difference between the confusing classes in the second-step modeling.

Feasibility of a Low-Power, Low-Resolution Laser-Induced Breakdown Spectroscopy Instrument for Analysis of Nickel Alloys: Quantification of the Major Alloying Elements and Classification

“…The two-step approach was previously applied to the sea salt classification problem with partial least squares discriminant analysis and found to be effective in increasing the classification accuracy particularly for the classes with minute differences. 33 The two-step LDA model was trained first to discriminate N1, N2, N5, N6, and the union of N3 and N4, and then the difference between N3 and N4 was separately modeled in the second stage. In this way, the minute difference between N3 and N4 could be more effectively exploited for modeling.…”

Atomic spectrometry update: review of advances in the analysis of clinical and biological materials, foods and beverages

“…Additionally, the spectral lines Al( i ) (394.424 nm), Al( i ) (396.152 nm), Ca( i ) (422.672 nm) and P( i ) (253.561 nm) were highly ranked and contributed significantly to the model. In a similar approach, Park et al 114 analysed sea salt from Japan, Korea and France, establishing Ca, K and Mg as the key features for classification. It was observed that those with higher levels of Mg also tended to have higher signals for H( i ) at 656 nm due to the absorption of water linked with Mg salts such as MgSO 4 and MgCl 2 .…”

Atomic spectrometry update: review of advances in the analysis of metals, chemicals and materials

“…In one by Park et al LIBS analysis was followed by a two-stage partial least squares discriminant analysis (PLSDA) to classify six commercial edible sea salt samples from three different countries (Japan, South Korea and France). 133 Although this is a fairly common approach during classification studies, this one had some novelty. In this study, the data were first sorted by the PLSDA into four classes and one extended class.…”