Examples of Stylometry Methods and How to Circumvent Them

For this example I’m going to use an essay by George Orwell, titled The Atom Bomb and You. Analyzing this text using the n-gram model method, there are a few notable examples:

  • turned out to be (3)
  • at the same time (2)
  • on the other hand (2)
  • has turned out to (2)
  • the dominant weapon is (2)
  • the invention of the (2)
  • the world between them (2)

Once a number of these data sets have been gathered, a linguist could potentially start to build profiles in which they can use to examine other types of writings that feature similar wording, grammar, punctuation, etc. There are of course limitations to this methodology. For the most accurate results the source would ideally be an essay, or personal exchanges of communications. The issues with nonfiction writings is they may drastically vary in terms of language and style.