Stylometry for E-mail Author Identification and Authentication
K. Calix, Mary M. Connors, David Levy, H. Manzar, Gavin McCabe, Sandy Westcott · 2008
The identification of the authorship of e-mail messages is of increasing importance due to an increase in the use of e-mail for criminal purposes. An author’s unique writing style can be reduced to a pattern by making measurements of various stylometric features from the written text. This paper reports on work to optimize and extend an existing C # based stylometry system that identifies the author of an arbitrary e-mail by using fifty-five writing style features. The program has been extended to provide feature vector data in a format appropriate for distribution to other project teams for subsequent data mining and classification experiments.