Extended Abstract: Predicting the Morality of a Character Using Character-Centric Embeddings

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Story generation and analysis have been researched for decades using AI and Natural Language Processing technology. However, while ethics is becoming essential for developing AI applications, little research deals with morality in the narrative. We present the framework to build embeddings for representing characters in terms of the character's morality. In addition, we propose a morality judgment task using storybooks. We conduct experiments, and the results suggest that word embedding models can learn a character's morality.

Original languageEnglish
Title of host publicationProceedings - 2023 IEEE International Conference on Big Data and Smart Computing, BigComp 2023
EditorsHyeran Byun, Beng Chin Ooi, Katsumi Tanaka, Sang-Won Lee, Zhixu Li, Akiyo Nadamoto, Giltae Song, Young-guk Ha, Kazutoshi Sumiya, Wu Yuncheng, Hyuk-Yoon Kwon, Takehiro Yamamoto
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages345-346
Number of pages2
ISBN (Electronic)9781665475785
DOIs
StatePublished - 2023
Event2023 IEEE International Conference on Big Data and Smart Computing, BigComp 2023 - Jeju, Korea, Republic of
Duration: 13 Feb 202316 Feb 2023

Publication series

NameProceedings - 2023 IEEE International Conference on Big Data and Smart Computing, BigComp 2023

Conference

Conference2023 IEEE International Conference on Big Data and Smart Computing, BigComp 2023
Country/TerritoryKorea, Republic of
CityJeju
Period13/02/2316/02/23

Keywords

  • Masked Language Modeling
  • Morals
  • Story Character Embedding

Fingerprint

Dive into the research topics of 'Extended Abstract: Predicting the Morality of a Character Using Character-Centric Embeddings'. Together they form a unique fingerprint.

Cite this