Whodunnit?: An Introduction to Practical Text Mining
How can one start to identify the author of an anonymous or collaborative text? Whether it be a legal document, a manifesto, or a supposedly authorless novel, digital stylometry can provide us an inroad into further research. In this workshop, we’ll explain how authorship software works, how to responsibly build a corpus for an authorship study, and how to work around its limitations (such as deliberately imitative texts). These topics will be organized around a case study of sequels to Charles Dickens’s unfinished novel, The Mystery of Edwin Drood.