Since Fall 2019, I have been a Co-PI of a research and educational project on “Shakespeare and Artificial Intelligence,” which started and developed as a collaboration with the Watt Family Center. The project team includes myself, Dr. Carl Ehrett (Watt), as well as two student assistants who have worked on the project for multiple semesters. Our goal is to use Natural Language Processing (NLP) tools to analyze the plays of Shakespeare and his contemporaries for educational and scholarly research. We have built and developed our own AI customized model to study analyze emotions like compassion and aggression, group dynamics and intergroup relationships, tensions, and discrepancies in Shakespeare’s comedies, tragedies, and histories (e.g. interactions between dominant groups and outsiders). This model generates sentiment and emotion scores that describe in quantitative terms the emotions expressed by the text. We then aggregate this sentiment/emotion data by character categories (e.g. male vs. female, social outsider vs. insider), fit classifier models that can distinguish genres of plays and/or between categories of characters.
Our project has a strong pedagogical aspect insofar as it incorporates a creative inquiry aspect:
1) Gather undergraduate students interested in AI/ML research;
2) Weekly project team meeting with both subject domain expert and with AI guide;
3) Students assist in applying AI/ML methods to research questions.
In addition, our project has resulted in the following important outcomes:
1) four student poster presentations as part of the annual Watt Clemson Student Research Forum and the Focus on Creative Inquiry.
2) two research presentations (“Using AI to Understand Shakespeare’s Drama”), in which Dr. Ehrett and I gave an overview of the history, methods, and findings of our project (2021 Annual Clemson Research Symposium)
3) Angela Nixon’s comprehensive feature article detailing the objectives, methods, and outcomes of the project appeared in Clemson World last summer. (https://clemson.world/line-by-line/)
4) we have completed a research paper, “Shakespeare Machine: New AI-Based Technologies for Textual Analysis,” which has been submitted for peer review in one of the world’s top-tier Digital Humanities journals, Digital Scholarship in the Humanities.
LUCIAN GHITA 