Dungeons and Dragons is a tabletop roleplaying game which focuses heavily on character interaction and creating narratives. The current state of the game's character creation process often bogs down new players in decisions related to game mechanics, not a character's identity and personality. This independent study investigates the use of machine learning and natural language processing to make these decisions for a player based on their character's backstory - the textual biography or description of a character. The study presents a collection of existing characters and uses these examples to create a family of models capable of predicting a character's class, race, and attribute scores. The accuracy and limitations of these models are discussed, but they represent a first step in abstracting away some of the more tedious parts of character creation for new players.


Sommer, Nathan


Computer Science


Artificial Intelligence and Robotics | Numerical Analysis and Scientific Computing


machine learning, dungeons and dragons, natural language processing

Publication Date


Degree Granted

Bachelor of Arts

Document Type

Senior Independent Study Thesis Exemplar



© Copyright 2019 Joseph C. MacInnes