The Use of Stylometry for Email Author Identification: A Feasibility Study
Robert Goodman, Matthew W. Hahn, Madhuri Marella, Christina Ojar, Sandy Westcott · 2007
It is recognized that authors have unique writing styles in which it is possible to find a simple statistical model. This paper makes an attempt to explain the methodology of developing and optimizing a stylometry program which quantifies the writing styles of various authors using 62 stylistic features found in emails. The feature extraction and pattern classification program was developed for a Pace University SSCSIS Doctorate of Professional Studies student completing her thesis on stylometry. The C#-based system is used to identify the author of an arbitrary email using stylometry features. Raw keystroke data, collected from an Internet-based Java applet in an earlier keystroke biometric study conducted by Dr. Mary Valliani, was converted into simple text files, appropriate features were extracted and a pattern classifier was implemented.