A novel normalization method based on principal component analysis to reduce the effect of peak overlaps in two-dimensional correlation spectroscopy

Enhanced 3-D asynchronous correlation data preprocessing method for Raman spectroscopy of Chinese handmade paper

We have developed a novel 3D asynchronous correlation method (3D-ACM) designed for the classification and identification of Chinese handmade paper samples using Raman spectra and machine learning. The 3D-ACM approach involves two rounds of tensor product and Hilbert transform operations. In the tensor product process, the outer product of the spectral data from different samples within the same category is computed, establishing inner connections among all samples within that category. The Hilbert transform introduces a 90-degree phase shift, resulting in a true three-dimensional spectral data structure. This expansion significantly increases the number of equivalent frequency points and samples within each category. This enhancement substantially boosts spectral resolution and reveals more hidden information within the spectral data. To maximize the potential of 3D-ACM, we employed six machine learning models: principal component analysis (PCA) with linear regression (LR), support vector machine (SVM) with LR, k-Nearest Neighbors (KNN), random forest (RF), and convolutional neural network (CNN). When applied to the 3D-ACM data preprocessing method, R-squared values of PLS-LR, KNN, RF and CNN supervised models, approached or equaled 1. This indicates exceptional performance comparable to unsupervised models like PCA. 3D-ACM stands as a versatile mathematical technique not confined to spectral data. It also eliminates the necessity for additional experimental setups or external control conditions, distinct from traditional two-dimensional correlation spectroscopy. Moreover, it preserves the original experimental data, setting it apart from conventional data preprocessing methods. This positions 3D-ACM as a promising tool for future material classification and identification in conjunction with machine learning.

Spectral insight into intensity variations in phase-transition processes using two-dimensional correlation analysis

Two-dimensional correlation spectroscopy (2D-COS) is widely used in studies of phase-transition processes to provide valuable order information that is useful in investigating these mechanisms. During phase-transition processes, the spectral intensity always changes in an "S"- or "anti-S"-shaped curve. Sample selection of the large and complicated dataset for 2D correlation analysis may impact the resulting sequence and should be given serious consideration. Additionally, the relationship between the sequential order obtained from 2D-COS and the parameters of the intensity change, namely, the transition point and the change rate, is still poorly defined. This article makes an attempt to resolve these problems based on the analysis of simulated spectra by assuming that the band intensity changes in a sigmoid manner without a step delay. It is concluded that the sample range around a transition point with a drastic intensity change defined by asynchronous perturbation-correlation moving-window (PCMW2D) analysis is a reasonable choice, and in this region a band that changes earlier as determined by 2D-COS most likely has an earlier phase-transition point. Also, from the results of segmental analysis, it is proposed that 2D-COS can distinguish the sequence of two bands using only rate difference; however, the rate difference and the form of the intensity change should be considered comprehensively. The insights from the simulated results are applied to analyze the temperature-dependent infrared (IR) spectra of poly[di(butyl)vinyl terephthalate] (PDBVT). The phase-transition mechanism of PDBVT can be clearly found using a suitable sample selection method.

An insight into sequential order in two-dimensional correlation spectroscopy

Determination of the sequential order of events is a very important feature of generalized two-dimensional correlation spectroscopy (2D-COS). Recently, queries have been put forward on the actual effectiveness of this method when changes are non-periodic, particularly in the presence of local sequential order. Consequently, it brings an urgent necessity to understand the true connotation of the sequential order parameter derived from 2D-COS analysis. This article presents an attempt to address these questions based on the analysis of simulated spectra by assuming the band intensity changes in a logarithmic, exponential, hyperbolic, or polynomial manner. It is concluded that for two events changing monotonically and without local sequential orders, one event occurring prior to the other as determined using 2D-COS means in most cases that the former has both a shorter half-time and a greater half-intensity. As a rule of thumb, intensity versus the perturbation factor should be plotted before performing the 2D-COS analysis to determine the sequential order of the involved events. In the presence of obvious local sequential order, 2D-COS analysis is unnecessary. Otherwise, sequential order can be determined quite reliably based on 2D-COS analysis.