a) obtaining a profile model of the structure, the profile model being defined by profile parameters that characterize the geometric shape of the structure;

b) obtaining a training set of values for the profile parameters;

c) generating a training set of simulated diffraction signals using the training set of values for the profile parameters, each simulated diffraction signal characterizing the behavior of light diffracted from the structure, wherein the simulated diffraction signals in the training set of simulated diffraction signals are defined using a standard set of signal parameters comprising a reflectance parameter, which characterizes the change in intensity of light when reflected on the structure, and a polarization parameter, which characterizes the change in polarization states of light when reflected on the structure, wherein the polarization parameter comprises:

a first polarization parameter that characterizes a difference between the square of the absolute value of complex reflection coefficients average over depolarization effects, and normalized to the reflectance parameter;

a second polarization parameter that characterizes an imaginary component of an interference of the complex reflection coefficients average over depolarization effects, and normalized to the reflectance parameter; and

a third polarization parameter that characterizes a real component of an interference of the complex reflection coefficients average over depolarization effects, and normalized to the reflectance parameter;

d) training a support vector machine using the training set of values for the profile parameters as inputs to the support vector machine and the training set of simulated diffraction signals as expected outputs of the support vector machine;

e) obtaining a measured diffraction signal off the structure, wherein the measured diffraction signal is defined using the standard set of signal parameters;

f) generating a simulated diffraction signal using a set of values for the profile parameters as inputs to the trained support vector machine, wherein the simulated diffraction signal is defined using the standard set of signal parameters;

g) comparing the measured diffraction signal to the simulated diffraction signal generated in f); and

h) when the measured diffraction signal and simulated diffraction signal match within one or more matching criteria, determining values of profile parameters of the structure to be the set of values for the profile parameters used in f) to generate the simulated diffraction signal.