Abstract
Recently, there has been increased interest in utilizing characters or subwords for natural language processing (NLP) tasks. However, the effect of utilizing character, subword, and word-level information simultaneously has not been examined so far. In this paper, we propose a model to leverage various levels of input features to improve on the performance of an supersense tagging task. Detailed analysis of experimental results show that different levels of input representation offer distinct characteristics that explain performance discrepancy among different tasks.- Anthology ID:
- W17-4106
- Volume:
- Proceedings of the First Workshop on Subword and Character Level Models in NLP
- Month:
- September
- Year:
- 2017
- Address:
- Copenhagen, Denmark
- Editors:
- Manaal Faruqui, Hinrich Schuetze, Isabel Trancoso, Yadollah Yaghoobzadeh
- Venue:
- SCLeM
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 41–45
- Language:
- URL:
- https://aclanthology.org/W17-4106
- DOI:
- 10.18653/v1/W17-4106
- Cite (ACL):
- Youhyun Shin and Sang-goo Lee. 2017. Supersense Tagging with a Combination of Character, Subword, and Word-level Representations. In Proceedings of the First Workshop on Subword and Character Level Models in NLP, pages 41–45, Copenhagen, Denmark. Association for Computational Linguistics.
- Cite (Informal):
- Supersense Tagging with a Combination of Character, Subword, and Word-level Representations (Shin & Lee, SCLeM 2017)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/W17-4106.pdf