Date of Graduation
Master of Science in Computer Science (MS)
Computer Science & Computer Engineering
Second Committee Member
Data Analysis, Data Visualization, Information Retrieval, Shakespeare Social Network, Social Network Analysis, Text Mining
With the emergence of digitization, large text corpora are now available online that provide humanities scholars an opportunity to perform literary analysis leveraging the use of computational techniques. This work is focused on applying network theory concepts in the field of literature to explore correlations between the mathematical properties of the social networks of plays and the plays’ dramatic genre, specifically how well social network metrics can identify genre without taking vocabulary into consideration. Almost no work has been done to study the ability of mathematical properties of network graphs to predict literary features. We generated character interaction networks of 36 Shakespeare plays and tried to differentiate plays based on social network features captured by the character network of each play. We were able to successfully predict the genre of Shakespeare’s plays with the help of social network metrics and hence establish that differences of dramatic genre are successfully captured by the local and global social network metrics of the plays. Since the technique is highly extensible, future work can be extended for fast and detailed literary analysis of larger groups of plays, including plays written in different languages as well as plays written by different authors.
Shukla, Manisha, "Theatrical Genre Prediction Using Social Network Metrics" (2018). Theses and Dissertations. 2920.