Our methodology is based on the CHvibration region on which a
set of spectral treatments was applied, from baseline subtraction to
spectral decomposition. Multivariate analyses were then done on
the treated data to differentiate into clusters the organic natural
materials. Studies involving Principal Component Analysis (PCA) on
diverse spectroscopic data were already done on different types of
samples and with different aims: material recognition [8,13,31,32]
or identification of acompound in a mixture [33]. We here focused
on the ability of these analyses to cluster samples with similar
vibrational characteristics. Since these spectroscopic features refer
to particular vibrations in specific molecular environments, new
discriminating criteria for material recognition were established.