Abstract
This paper describes our submission to the CoNLL-2015 shared task on discourse parsing. We factor the pipeline into sub-components which are then used to form the final sequential architecture. Focusing on achieving good performance when inferring explicit discourse relations, we apply maximum entropy and recurrent neural networks to different sub-tasks such as connective identification, argument extraction, and sense classification. The our final system achieves 16.51%, 12.73% and 11.15% overall F1 scores on the dev, WSJ and blind test sets, respectively.
Original language | English |
---|---|
Title of host publication | CoNLL 2015 - 19th Conference on Computational Natural Language Learning, Proceedings of the Shared Task |
Publisher | Curran Associates Inc. |
Pages | 89-94 |
Number of pages | 6 |
ISBN (Electronic) | 1932432671, 9781932432671 |
Publication status | Published - 2014 |
Externally published | Yes |
Event | 19th Conference on Computational Natural Language Learning: Shared Task, CoNLL 2015 - Beijing, China Duration: 30 Jul 2015 → 31 Jul 2015 |
Publication series
Name | CoNLL 2015 - 19th Conference on Computational Natural Language Learning, Proceedings of the Shared Task |
---|
Conference
Conference | 19th Conference on Computational Natural Language Learning: Shared Task, CoNLL 2015 |
---|---|
Country/Territory | China |
City | Beijing |
Period | 30/07/15 → 31/07/15 |
Cite this
Wang, L., Hokamp, C., Okita, T., Zhang, X., & Liu, Q. (2014). The DCU discourse parser for connective, argument identification and explicit sense classification. In CoNLL 2015 - 19th Conference on Computational Natural Language Learning, Proceedings of the Shared Task (pp. 89-94). (CoNLL 2015 - 19th Conference on Computational Natural Language Learning, Proceedings of the Shared Task). Curran Associates Inc..