A Computational Analysis of the Voices of Shakespeare’s Characters

Proceedings of Recent Advances in Natural Language Processing, pages 295–300 Varna, Sep 4–6, 2023 https://doi.org/10.26615/978-954-452-092-2_033 295 A Computational Analysis of the Voices of Shakespeare’s Characters Liviu P. Dinu, Ana Sabina Uban University of Bucharest, Romania ldinu@fmi.unibuc.ro, auban@fmi.unibuc.ro Abstract In this paper we propose a study of a rela- tively novel problem in authorship attribution research: that of classifying the stylome of characters in a literary work. We choose as a case study the plays of William Shakespeare, presumably the most renowned and respected dramatist in the history of literature. Previous research in the field of authorship attribution has shown that the writing style of an author can be characterized and distinguished from that of other authors automatically. The ques- tion we propose to answer is a related but differ- ent one: can the styles of different characters be distinguished? We aim to verify in this way if an author managed to create believable charac- ters with individual styles, and focus on Shake- speare’s iconic characters. We present our ex- periments using various features and models, including an SVM and a neural network, show that characters in Shakespeare’s plays can be classified with up to 50% accuracy. 1 Introduction The problem of authorship identification is based on the assumption that there exist stylistic features that can help distinguish the real author of a text from any other theoretical author, and that these can be computationally measured and exploited in order to automatically identify the true author of a text. Automated authorship attribution has a long and rich history (starting from the early 20th century (Mendenhall, 1901)) and has since then been extensively studied and elaborated upon. One of the most influential studies in authorship attribution is the study of (Mosteller and Wallace, 1963) on the Federalist Papers, in which the authors try to determine the real author of a few of these papers which have disputed paternity. In this work, they both introduce a standard dataset and propose an effective method for distinguishing between the author’s styles, based on function words frequen- cies, that is still relevant and used to this day. Many types of features have been proposed and success- fully used in subsequent studies to determine the author of a text. These types of features generally contrast with the content words commonly used in text categorization by topic, and are said to be used unconsciously and harder to control by the author. Such features are, for example, function words (Mosteller and Wallace, 1963; Dinu et al., 2012), grammatical structures (Baayen et al., 1996), part-of-speech n-grams (Koppel and Schler, 2003), lexical richness (Tweedie and Baayen, 1998), or even the more general feature of character n-grams (Keˇselj et al., 2003; Dinu et al., 2008). Recent stud- ies focusing on stylistic variation within the writ- ings of a single author combine traditional function word features with stylistic markers such as lexical richness and readability, as well as topic modelling, to compare the importance of the the stylome and the topics discussed in …

Content truncated to 500 words