Introduction

My name is Helena Aleluya José, and I’m an international student from Angola. I’m also a Computer Science and Theater double major, and an African and African-American Studies and Creative Writing double minor. As both a computer scientist and a playwright, I wanted to create a project that reflects the intersection of my passions. For my senior capstone, I developed a framework that applies topic modeling techniques (LDA) to a corpus of dramatic texts in order to study playwright originality

Abstract

This capstone project aims to develop a novel approach for analyzing and comparing playwrights’ unique voices and styles by applying topic modeling techniques, specifically Latent Dirichlet Allocation (LDA), on a corpus of theater texts. By leveraging advanced natural language processing (NLP) methods, the project will preprocess and prepare a diverse collection of plays for topic modeling, implement custom algorithms tailored for dramatic literature, and analyze the discovered topics to identify recurring themes, character archetypes, and narrative structures prevalent across different playwrights’ works. Interactive visualization tools will be developed to facilitate the exploration and interpretation of these insights, enabling literary scholars and critics to understand the creative processes better and the influences that shape dramatists’ voices.

Keywords 

Topic modeling techniques, textual analysis, LDA model, Digital Humanities