International Journal of Computational Linguistics Research Edoardo Saccenti, Leonardo Tenori 3 2 2012 http://www.dline.info/jcl/fulltext/v3n2/1.pdf We analyzed the three parts of Dante’s Divina Commedia using word frequencies as style markers for statistical analysis. Partial least square discriminant analysis was used to provide separation among the cantos of the three parts of the Comedy, Inferno, Purgatorio and Paradiso; the statistical models were able to successfully discriminate the poetic tones used by Dante to characterize and diversify the three parts, demonstrating the existence of a particular stylistic substratum in each cantica of the Commedia.