Authorship analysis has a background in stylometry, which is the study of an author's style of writing. The concept is based on the idea that everyone learns language slightly differently, and that measuring these nuances in people's writing will enable us to tell them apart using only the content of their writing.

Authorship analysis has historically (pre-1990) been performed using repeatable manual analysis and statistics, which is a good indication that it could be automated with data mining. Modern authorship analysis studies are almost entirely data mining-based, although quite a significant amount of work is still done with more manually driven analysis using linguistic styles and stylometrics. Many of the advances in feature engineering today are driven by advances in stylometrics. In other words, manual analysis discovers new features, which are then codified and used as part of the data mining process.

A key underlying feature of stylometry is that...