CODICES Digital Humanities Lab

Dates

2019 - Present

Project Abstract

Aristotle famously described tragedy as a literary form that evokes feelings of pity and fear in its audience that lead to a sense of purification or catharsis. This project asks the extent to which the computational techniques of sentiment analysis and topic modeling can be used to identify and map these emotions on the collection of tragedies written by Aeschylus, Sophocles, and Euripides.

Our current project builds on previous work that constructed visualization tools allowing for the exploration of linguistic data in Greek tragedy using social networks integrated with linguistic data. This work is available online at https://daedalus.umkc.edu/VisualExplorer/, and it was published in the Proceedings of the 2011 Chicago Colloquium on Digital Humanities.

We have published two articles on this work that explore how human readers understand the sentiments of passages from the text and how this compares to computational sentiment analysis. Our goal has been to validate the applicability of computational sentiment analysis methods designed for Tweets and Facebook posts to literary texts.

Publications

• "Interpretation of Sentiment Analysis in Aeschylus's Greek Tragedy," Vijaya Kumari Yeruva, Mayanka Chandrashekar, Yugyung Lee, Jeff Rydberg-Cox, Virginia Blanton, Nathan A Oyler. Published in Proceedings of The 4th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature (December 2020). Available at https://www.aclweb.org/anthology/2020.latechclfl-1.17/.
• "Interpretation of Sentiment Analysis with Human-in-the-Loop," Vijaya Kumari Yeruva, Mayanka Chandrashekar, Yugyung Lee, Jeff Rydberg-Cox, Virginia Blanton, Nathan A Oyler. Forthcoming in Proceedings of The 4th IEEE Workshop on Human-in-the-Loop Methods and Future of Work in BigData (December 2020). Available at https://humanmachinedata.org.

Our current work focuses on computationally identifying main topic clusters from the corpus of Greek tragedy, assigning sentiments to these topic clusters, and associating these topic-sentiment pairs with characters and character types in the plays.

Project Team