MBTI Personality Prediction for Fictional Characters Using Movie Scripts
Document Type
Conference Proceeding
Publication Date
1-1-2022
Abstract
An NLP model that understands stories should be able to understand the characters in them. To support the development of neural models for this purpose, we construct a benchmark, Story2Personality. The task is to predict a movie character's MBTI or Big 5 personality types based on the narratives of the character. Experiments show that our task is challenging for the existing text classification models, as none is able to largely outperform random guesses. We further proposed a multi-view model for personality prediction using both verbal and non-verbal descriptions, which gives improvement compared to using only verbal descriptions. The uniqueness and challenges in our dataset call for the development of narrative comprehension techniques from the perspective of understanding characters.
Identifier
85149829756 (Scopus)
ISBN
[9781959429432]
Publication Title
Findings of the Association for Computational Linguistics Emnlp 2022
External Full Text Location
https://doi.org/10.18653/v1/2022.findings-emnlp.448
First Page
6744
Last Page
6753
Recommended Citation
Sang, Yisi; Mou, Xiangyang; Yu, Mo; Wang, Dakuo; Li, Jing; and Stanton, Jeffrey, "MBTI Personality Prediction for Fictional Characters Using Movie Scripts" (2022). Faculty Publications. 3521.
https://digitalcommons.njit.edu/fac_pubs/3521
