$a$:	Depth of cut (mm)
$ACC$:	Resultant PSD distribution
$AC{C}_x$:	PSD distribution of x-axis accelerometer signal
$AC{C}_y$:	PSD distribution of y-axis accelerometer signal
$AC{C}_z$:	PSD distribution of z-axis accelerometer signal
BPNN:	Backpropagation neural network
$C$:	Covariance matrix of the training set
$F$:	Calculated cutting force (N)
$f$:	Cutting feed (mm/rev)
${\gamma }_0$:	Rake angle
$h_m$:	Chip thickness (mm)
$k_c$:	Specific cutting force (N/mm2)
$k_{c1}$:	Specific cutting force (N/mm2)
$K_r$:	Edge angle
${\lambda }_i$:	Eigenvalue
$m_c$:	Non-dimensional factor
PCA:	Principal component analysis
$P{I}_i$:	Weight value
PSD:	Power spectral density
$q$:	Ratio of the total variance
$S$:	Training set
$S_{ij}$:	Insert condition classification
$S_B$:	Weighted sum value for the built-up edge insert condition
$S_F$:	Weighted sum values for the fracture insert condition
$S_N$:	Weighted sum value for the normal insert condition
$S_W$:	Weighted sum value for the flank wear insert condition
$u$:	Unitless coefficient
$v_i$:	Eigenvector
$W$:	Transformation matrix
$x$:	Test data
$x_i$:	Training data
$\overline{X}$:	Average of the training set
$y$:	Transformed data