Abstract
We consider the task of detecting contractual obligations and prohibitions. We show that a self-attention mechanism improves the performance of a BILSTM classifier, the previous state of the art for this task, by allowing it to focus on indicative tokens. We also introduce a hierarchical BILSTM, which converts each sentence to an embedding, and processes the sentence embeddings to classify each sentence. Apart from being faster to train, the hierarchical BILSTM outperforms the flat one, even when the latter considers surrounding sentences, because the hierarchical model has a broader discourse view.- Anthology ID:
- P18-2041
- Volume:
- Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
- Month:
- July
- Year:
- 2018
- Address:
- Melbourne, Australia
- Venue:
- ACL
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 254–259
- Language:
- URL:
- https://aclanthology.org/P18-2041
- DOI:
- 10.18653/v1/P18-2041
- Cite (ACL):
- Ilias Chalkidis, Ion Androutsopoulos, and Achilleas Michos. 2018. Obligation and Prohibition Extraction Using Hierarchical RNNs. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 254–259, Melbourne, Australia. Association for Computational Linguistics.
- Cite (Informal):
- Obligation and Prohibition Extraction Using Hierarchical RNNs (Chalkidis et al., ACL 2018)
- PDF:
- https://preview.aclanthology.org/remove-xml-comments/P18-2041.pdf