Keyness in Shakespeare's Plays

This paper describes a novel method of identifying keyness in texts using relational learning. The relational learning is based on Inductive Logic Programing (ILP) in first-order logic. Using this method we can extract first-order logic formulas satisfied in a set of positive examples and not satisfied in a set of negative examples. We tested this method on a collection of Shakespeare’s plays to identify keyness (or aboutness) of particular plays. The research was especially related to Hamlet, Prince of Denmark which was already investigated by Mike Scott [0]. He used his own tool WordSmith, based on wordlists generating. Aim of this paper is to describe another way of automatic identifying keyness and to show that this method can findmore comprehensive keyness representation.