Enhancing counterfeit and illicit medicines grouping via feature selection and X-ray fluorescence spectrometry.

•We propose a new feature selection approach for XRF data based on multidimensional scaling (MDS) and Procrustes analysis (PA).•Our propositions are applied to 11 authentic commercial brands and seized counterfeit samples.•The proposed framework improved the stratification of samples.•Results support investigative forces towards tracking and interrupting counterfeit operations.