Effects of scatter-correction pre-processing methods and spectral derivative algorithms on forensic classification of paper

Abstract

Abstract. Infrared (IR) spectral data are always influenced by undesired random and systematic variations. As such, preprocessing of spectral data is normally required before chemometric modeling. Two most widely used pre-processing techniques, i.e. scatter-correction methods and spectral derivatives, were used to pre-process 150 IR spectral data of paper. The algorithms investigated in this preliminary study are Standard Normal Variate (SNV), Multiplicative Scatter Correction (MSC), Savitzky-Golay (SG) and Gap-Segment (GS). The visual examination of the clustering among three studied varieties of paper, i.e. IK Yellow, One Paper and Save Pack, is accomplished via Principal Component Analysis (PCA). Overall, separation of the three varieties of paper is greatly enhanced after pre-processing. The most significant improvement is obtained with pre-processing via 1st derivative using SG algorithms.

Partial least squares-discriminant analysis (PLS-DA) for classification of high-dimensional (HD) data: a review of contemporary practice strategies and knowledge gaps

“…Mathematically, PLS-DA modelling is not a one-step procedure but involves a series of mathematical operations and a wealth of parameters. It is the first author's experience who is enthusiastic about the potential of PLS-DA in modelling infrared (IR) spectra for solving forensic-based problems, [15][16][17] but finds no work addressing systematically the general PLS-DA modelling practice strategies. Although some papers have described and critically discussed the pitfalls of PLS-DA, 2,6,[18][19][20][21] we noticed that the decision rule (DR) and empirical differences between PLS1-DA versus PLS2-DA algorithms have not been elaborated in detail but only briefly discussed on the theoretical ground.…”