Fourier transform and near infrared dataset of dialdehyde celluloses used to determine the degree of oxidation with chemometric analysis
Description
All data refer to the original research article “A fast method to measure the degree of oxidation of dialdehyde celluloses using multivariate calibration and infrared spectroscopy”. 1) The developed NIR/PLSR models (1 and 2) and FTIR/PLSR models (3 and 4) with the corresponding spectra are provided: “Dataset”>”processed_data” (.q2 and .0 files) 2) In addition, the spectral raw data for each model (1-4) is provided, if a different chemometrics software is used: "Dataset">"raw_data.zip" (.csv files)
Files
Steps to reproduce
There are two ways to reproduce the NIR/PLSR and FTIR/PLSR models. The easiest way would be to use Quant2 Bruker OPUS software package. In this case, the models can be just used as-is ("processed_data"). If other software is used to reproduce the developed models, the spectral data is also attached as CVS. files (raw_data.zip). The corresponding cross-validation results and parameters of partial least-squares regression are summarized in the manuscript. The models can be quickly rebuilt with any chemometrics software with the spectral data and the parameters in hand.