Determining the Gender of Shakespeare’s Characters

The purpose of this project was to determine the gender of characters in Shakespeare’s plays by applying machine learning algorithms to the characters’ word usage. In particular, a naive Bayes model and support vector machine model were implemented on a corpus consisting of the words spoken by 198 characters – 99 male and 99 female. Both models achieved reasonably good classification accuracy, with SVM unsurprisingly outperforming naive Bayes, 76.3% to 69.2%. Interestingly, characters from tragedies were classified with much greater accuracy than were characters from either comedies or histories.