/JDTB-logic

Japanese Discourse Relation Treebank focusing on Concessive Inferences

Creative Commons Attribution Share Alike 4.0 InternationalCC-BY-SA-4.0

Japanese Discourse Relations Focusing on Concessive Relations

A repository for our LREC-COLING2024 paper "Annotation of Japanese Discourse Relations Focusing on Concessive Inferences."

Data

The fields in the files data/nagara.tsv, data/tsutsu.tsv, and data/tokorode.tsv are as follows:

  • kainoki_id: ID in the Kainoki Treebank
  • first_half: The first half of the sentence where the connective appears
  • connective: The target connective
  • second_half: The second half of the sentence where the connective appears
  • arg1: The first discourse relation unit (the first argument that appears in the clause that is syntactically bound to the connective)
  • arg2: The second discourse relation unit (the second argument that appears in the clause that is syntactically bound to the connective)
  • classification_label: Discourse relation label
  • For linguistic feature labels ASP, NEG, MOD, CNJ, FOC, QUE, IMP, see Section 4.1 of our paper.

Citation

If you use this data in any published research, please cite the following:

@inproceedings{kubota-etal-2024-annotation-japanese,
    title = "Annotation of {J}apanese Discourse Relations Focusing on Concessive Inferences",
    author = "Kubota, Ai  and
      Sato, Takuma  and
      Amamoto, Takayuki  and
      Akiyoshi, Ryota  and
      Mineshima, Koji",
    booktitle = "Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)",
    year = "2024",
    pages = "1215--1224"
}

Contact

For questions, please contact minesima@abelard.flet.keio.ac.jp.

License

CC BY-SA 4.0