Synapse
⌘+K
Synapse
PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
August 2, 2026ACM Transactions on Asian and Low-Resource Language Information ProcessingOpen Access

Large-Scale Word Sense Tagging in Contemporary Japanese: An All-Words Word Sense Disambiguation Approach for 180 Million Words

View Full Paper
Ask AI
Bookmark
Share

Authors

KKKanako KomiyaSASoma AsadaMAMasayuki Asahara

Discussion

Loading...

Member takes

Overview

Randomized trial demonstrates effective word sense disambiguation in multiple Japanese corpora, highlighting its application in language education.

Key Points

  • The aim is to develop a system for automatic word sense disambiguation in Japanese, enhancing corpus accessibility and educational potential.
  • Utilized a BERT-based model fine-tuned on the Balanced Corpus of Contemporary Written Japanese (BCCWJ).
  • Performed word sense tagging on various Japanese corpora with a focus on semantic principles.
  • Manually evaluated 500 words across different domains for accuracy.
  • Achieved accuracies of 68.2% to 91.8% for word sense tagging, varying by domain.
  • Developed Chunagon, a public application for searching the corpus based on word senses.

Cite This Study

Komiya et al. (2026) studied this question.

synapsesocial.com/papers/6a6eeaf51b0468a7eeab3a7ahttps://doi.org/10.1145/3837069
View Full Paper
Ask AI
Bookmark
Share