An Evaluation Dataset for Identifying Communicative Functions of Sentences in English Scholarly Papers

The dataset consists of three sets of tsv files. The usage is explained in our paper.

Description of tsv files

human_evaluation

0. ID
1. Targeted sentence [s0]
2. Correct choice [s1]
3. Wrong choice [s2]
4. Core FE for s0
5. Core FE for s1
6. Core FE for s2
7. Communicative function for s0 and s1
8. Communicative function for s2
9. Paper/sentence ID for s0
10. Paper/sentence ID for s1
11. Paper/sentence ID for s2
12. Accuracy of human annotation

sentences

0. Communicative function
1. The core FE
2. Sentence
3. Sentence ID (PaperID_SentID; identical to the ID in AASC)

Licence

This dataset is licensed under the Creative Commons BY-NC-SA 3.0. When you use the dataset, please cite our paper (see below).

This dataset uses ACL Anthology Sentence Corpus, which consists of papers retrieved from ACL Anthology.

Papers in the ACL repository

© 1979-2018 Association for Computational Linguistics

Licensed under the Creative Commons BY-NC-SA 3.0 (-2015) and Creative Commons BY 4.0 (2016-)

AASC: ACL Anthology Sentence Corpus

Licensed under the Creative Commons BY-NC-SA 3.0

Citation

Iwatsuki, K., Boudin, F., & Aizawa, A. (2020). An Evaluation Dataset for Identifying Communicative Functions of Sentences in English Scholarly Papers. In Proceedings of The 12th Language Resources and Evaluation Conference, 1712–1720.

@InProceedings{Iwatsuki2020LREC,
  author    = {Iwatsuki, Kenichi  and  Boudin, Florian  and  Aizawa, Akiko},
  title     = {An Evaluation Dataset for Identifying Communicative Functions of Sentences in English Scholarly Papers},
  booktitle      = {Proceedings of The 12th Language Resources and Evaluation Conference},
  month          = {May},
  year           = {2020},
  address        = {Marseille, France},
  publisher      = {European Language Resources Association},
  pages     = {1712--1720},
  url       = {http://www.lrec-conf.org/proceedings/lrec2020/pdf/2020.lrec-1.212.pdf}
}