DOI: https://doi.org/10.15368/theses.2020.148
Available at: https://digitalcommons.calpoly.edu/theses/2233
Date of Award
12-2020
Degree Name
MS in Computer Science
Department/Program
Computer Science
College
College of Engineering
Advisor
Franz Kurfess
Advisor Department
Computer Science
Advisor College
College of Engineering
Abstract
The study of personality types gained traction in the early 20th century, when Carl Jung's theory of psychological types attempted to categorize individual differences into the first modern personality typology. Iterating on Jung's theories, the Myers-Briggs Type Indicator (MBTI) tried to categorize each individual into one of sixteen types, with the theory that an individual's personality type manifests in virtually all aspects of their life. This study explores the relationship between an individual's MBTI type and various aspects of their writing style. Using a MBTI-labeled dataset of user posts on a personality forum, three ensemble classifiers were created to predict a user's personality type from their posts with the goal of outperforming existing research as well as outperforming the test-retest reliability of online questionnaire-based personality assessments. With the increasing amount of textual data available today, the creation of an accurate text-based personality classifier would allow for user experience designers and psychologists to better tailor their services for their users.