Computational Stylometry: An Introduction

Moore Institute, NUI Galway, 24 April 2014

This one-day workshop led by Dr Francesca Benatti (Open University) is a hands-on introduction to stylometric analysis using Stylo, a tool created for use in the statistical programming language R.

What is Computational Stylometry?Computational stylometry (or stylistics) is the study of how the stylistic features of texts can be measured through computer-aided statistical methods.

How is it used?Regularly used in the field of authorship attribution, Computational Stylometry has potential applications to a wide range of fields and disciplines that examine texts. In addition to authorship attribution, Computational Stylometry may be used to investigate relationships between texts in such areas as gender, genre, and chronology.

For example?J. K. Rowling was recently exposed as the author of the pseudonymous novel, The Cuckoo’s Calling, thanks to the Computational Stylometric work of Patrick Juola. A professor of computer science, Juola also runs a private consulting firm that applies stylometric techniques to disputed documents and in psychological profiling for the legal professions. Similar techniques have aided historical research by verifying the authenticity of letters written during the American Civil War, and sought to reveal how pronouns and other function words reveal personality traits and relationship roles.

What does the workshop involve?It will include an introduction to the history and theory of stylometry, and will focus on the practical application of the Stylo package. Alongside examples of Stylo’s use, participants will complete hands-on practical exercises with the tool. A set of sample texts will be supplied, but participants are free to bring their own texts in plain text or TEI XML format.

What is the schedule for the day?10.00-10.30 Introduction to Stylometry10.30-11.00 Installation of R and Stylo; settings and options11.00-11.30 The stylo function; hands-on practice11.30-12.00 Tea break12.00-13.00 The oppose function; hands-on practice13.00-14.00 Lunch14.00-14.30 Case study: authorship attribution and the Christabel review14.30-15.30 The classify function; hands-on practice15.30-16.00 Open session: practice / discussion / Q and A

I am interested, but am not sure this is for me…Computational stylometry can be either a means or an end. While is it used to identify authors andverify documents, it can also be employed for more exploratory purposes: for identifying patternsand sequences within a text or group of texts. If your work involves analysing texts, this workshopmay be useful to you.

Do I need to be a programmer to learn this?No. This workshop is led by and aimed at scholars in the Arts and Humanities. Some experience withcomputer-aided text-analysis will be a benefit, but nothing other than general digital literary isrequired.

Anything else I should know?Attendance is free. Places are limited on this workshop, however, so you are advised to register yourinterest as soon as possible. Attendees will be required to bring their own laptops: R and Stylo willnot work on mobile devices.

Ok, you’ve sold me. What do I do now?Contact Justin Tonra (justin.tonra@nuigalway.ie) to register your interest in participating. Youshould mention any previous experience you have had with text-mining or text-analysis, anddescribe how the workshop relates to your own work.