This work describes a new method for the rapid characterization of volatile compounds of cork wine stoppers by a dynamic headspace method coupled to mass spectrometry (DHS-MS). This technique does not require any pretreatment of the sample. The global volatile signature of the cork is evaluated without real chromatographic separation of its constituents. For data analysis, the mass fragments of each spectrum (50 < m/z <180 amu) were considered as potential descriptors of the composition of the cork sample headspace. Chemometrics methods (principal component analysis and partial least-squares regression) were applied to extract useful information by selecting the most significant mass fragments (m/z) that allowed a good classification of the cork samples studied. In the present work, geographical origin of cork samples could be successfully characterized and predicted.