Skip to Main Content

Stylometry Methods and Practices

A review of various stylometry methods and programs for the digital humanities.

Types of Projects

Typical projects usually fall under these four categories:

  1. An anonymous text with no indication of who the author is
  2. An anonymous text with two, three, or some other small number of suspected authors
  3. An anonymous text that is the result of a collaboration
  4. An anonymous text with one suspected author

But there are also exceptions! It's important that you take the time and think about the specific goal for your project. 

Hildegard of Bingen: Authorship and Stylometry

This documentary demonstrates the use of a statistical method called Principle Component Analysis (this can be done in R-stylo) to look at the possibly misattributed authorship of some key medieval texts.

Shakespeare and Double Falsehood

Using function words alongside other cognitive "tells", this project looks at a supposedly "lost" Shakespeare play to determine authorship.

Rowling and "Galbraith"

This project used JGAAP to determine J.K. Rowling as the author of Cuckoo's Calling out of a set of likely candidates.

Go Set A Watchman while we Kill the Mockingbird In Cold Blood

​This recent project uses R-stylo to perform a "rolling.classify" analysis on Harper Lee's new novel, Go Set A Watchman. This kind of analysis is used on collaborations or co-authored works. The goal was to see how much Truman Capote influenced Lee's novel.