Fast identification of biominerals by means of stand-off laser‐induced breakdown spectroscopy using linear discriminant analysis and artificial neural networks

A critical review of recent progress in analytical laser-induced breakdown spectroscopy

“…The comparisons are generally positive for LIBS, and the studies always mention that an added benefit of LIBS is that it can be made portable. A further development in this direction was made in a study [199] that suggested and tested the use of stand-off LIBS directly on an excavation site in an attempt to make archeological field work less damaging and time-consuming by quickly classifying the type of objects found.…”

Evaluation of supervised chemometric methods for sample classification by Laser Induced Breakdown Spectroscopy

“…Furthermore LDA does not take into account the difference in dispersion of each group and it is not appropriate if the variance structure is different between groups [22]. LDA has been applied to the analysis of LIBS data [16,23].…”

“…Furthermore, the model must be able to classify a sample that does not belong to any Chemometrics and Intelligent Laboratory Systems 146 (2015) 354-364 class as unassigned and not classify it as another class (high robustness). Different chemometric methods have been employed in LIBS data analysis to perform reliable classification for various types of samples [14][15][16][17][18], even at remote sites such as rock classification on Mars. However, none of these works includes a complete analysis of the performance of these methods evaluating sensibility, generalization ability and robustness with very similar samples.…”

Classification of Geological Samples Based on Soft Independent Modeling of Class Analogy Using Laser-Induced Breakdown Spectroscopy

“…To simplify the analysis, LIBS applications on geological materials have been proposed over the last two decades. Furthermore, multivariate preprocessing methods have been increasingly studied, including approaches based on principal component analysis (PCA) [12,13], partial least squares discriminant analysis (PLS-DA) [14], graph theory (GT) [15], independent component analysis (ICA) [16], and artificial neural networks (ANNs) [17,18]. Such methods consider the effect of redundant information and hence increase the efficiency of data analysis and prevent negligible fluctuations resulting from experimental conditions and instrumental instability [19].…”