PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
June 2, 2026International Symposium on Affective Science and EngineeringOpen Access

Inferring Emotion Words in Occupational Fatalities Using a Large Language Model

View Full Paper
Ask AI
Bookmark
Share

Authors

MSMasaru Sato

Discussion

Loading...

Member takes

Overview

This randomized trial analyzes emotion words in fatal accident narratives to inform industry-specific safety training.

Key Points

  • This research aims to uncover patterns of inferred emotional responses in narratives of occupational fatalities across industries.
  • Utilized a large language model to infer emotion words from narratives of occupational fatal accidents.
  • Categorized inferred emotions into nine groups and summarized their frequencies across manufacturing, construction, and transport/traffic.
  • Applied a chi-square test to analyze the association between industry and emotional category distribution.
  • Significant association found between industry and emotion distribution (χ²(16)=187.71, p<.001, Cramér’s V=0.143).
  • Tension and fear were overrepresented in the construction industry, while complacency and focus were dominant in manufacturing.
  • Fatigue and anxiety were most frequent in the transport/traffic sector compared to expected counts.

Cite This Study

Masaru Sato (2026) studied this question.

synapsesocial.com/papers/6a1e726230b38c64201b5ab7https://doi.org/10.5057/isase.2026-c000057
View Full Paper
Ask AI
Bookmark
Share