M	component matrix
X	vector of features
n	number of features
$W_m$	eigenvectors
k	number of scale features
$P_{sr}$	exposure index for a r-substances
$ND{S}_r$	the highest permissible concentration values of the substance
$\overline{r}$	average correlation coefficient between all pairs of items
$\dot{r}$	coefficient of determination
$\beta$	the criterion of the estimated number of features
$\Phi \left(W_m\right)$	standard normal distribution function $N(0,1)$
$\Omega$	original data space
$\Theta$	transformed data space
feature space D	teaching sample
K ($x,\beta$)	kernel function for PCA methods
GNB	Gaussian Naive Bayes
MLP	Multi-layer perceptron
k-NN	k Nearest Neighbours
SVC	Support Vector Classification/Support Vector Machine
CART	Classification and Regression Trees
NO	no feature extraction
SP	scree factor criterion
K	Kaiser criterion
${}^{\prime }$	Means the use of a proprietary method
PCA	Principal Component Analysis
CCPCA	Centroid Class Principal Component Analysis
GPCA	Gradient Stochastic Principal Component Analysis
KPCA	Kernel Principal Component Analysis