Character Space: using matrix factorization to identify main axes in fictional characters
Conference Year
January 2022
Abstract
Singular value decomposition is a powerful matrix reduction technique which we used to pull out core components of personality of fictional characters, yielding a character space. While these are interesting results in themselves, we also considered how they might map onto other spaces, particularly the power-danger-structure framework, and how this strategy could be extended to include other aspects of literary inquiry and other data.
Primary Faculty Mentor Name
Christopher M. Danforth
Secondary Mentor Name
Peter Dodds
Student Collaborators
Denis Hudon, Tyler Ehrlich, Philip Nguyen
Status
Graduate
Student College
College of Engineering and Mathematical Sciences
Program/Major
Complex Systems
Primary Research Category
Arts & Humanities
Secondary Research Category
Engineering & Physical Sciences
Character Space: using matrix factorization to identify main axes in fictional characters
Singular value decomposition is a powerful matrix reduction technique which we used to pull out core components of personality of fictional characters, yielding a character space. While these are interesting results in themselves, we also considered how they might map onto other spaces, particularly the power-danger-structure framework, and how this strategy could be extended to include other aspects of literary inquiry and other data.