Table 21.

Table of symbols

Symbol	Meaning
m	The number of features
$R_m$	The dimensional pattern space
c	The number of categories (i.e. classes)
$c_i$	A class at index i
${\omega }_i$	A class region at index i
$X_u$	An unknown pattern
f	A set of discriminant functions
$f_{c_i}$	A discriminant function at $c_i$
$c_k$	A class at index k
$\text{P}({\omega }_i|X_u)$	The posterior probability of finding $X_u$ in the class region ${\omega }_i$
$\text{P}(X_u|{\omega }_i)$	The probability that $X_u$ belongs to the class region ${\omega }_i$
$\text{P}({\omega }_i)$	The priori of a class region ${\omega }_i$
$\text{P}(X_u)$	The evidence of $X_u$
$n_{c_i}$	The number of samples in a class $c_i$
N	The total number of samples in the sample space
z	A measure of dependency on variables or the predicted output value
X	The variables or inputs
i	An index
k	An index
${\beta }_0$	The intercept or bias
${\beta }_i$	The slope of the logistic regression model or the coefficient value at index i
V	The number of variables or inputs
$\text{P}(z)$	The probability of the variables
l	The number of features (i.e., parameters) in $X_u$
j	An index
$\mu$	The mean value
$\sigma$	The standard deviation value
h	A smoothing parameter optimized on the training dataset
$x_{\mathit{vji}}$	The value of the feature in the jth position of the vth input in class $c_i$
$y_i$	The class number
w	The weight vector
b	The bias
C	A parameter
$L_{\mathit{hinge}}$	The hinge loss
TP	True positive
TN	True negative
FP	False positive
FN	False negative