Abstract | ||
---|---|---|
Supervised event extraction methods suffer from the lack of high-quality event corpora. Active learning is applied to improve the efficiency of manual annotation. In particular, we introduce the uncertainty of argument classification into the active learning for pipeline and joint extraction models. For the pipeline model, we drive active learning to identify and annotate the most informative instances at each extraction stage. It proceeds step-by-step and iteratively until the extraction at each stage reaches the optimal state. While for the joint model, we incorporate active learning with structural perceptron to identify the informative and interdependent event constituents. Experiments on ACE 2005 English corpora show that active learning for pipeline and joint model yield promising improvement. |
Year | DOI | Venue |
---|---|---|
2016 | 10.1007/978-981-10-2993-6_11 | Communications in Computer and Information Science |
Keywords | DocType | Volume |
extraction,Active learning,Structural perceptron | Conference | 669 |
ISSN | Citations | PageRank |
1865-0929 | 0 | 0.34 |
References | Authors | |
0 | 7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Kai Wang | 1 | 0 | 0.34 |
Yingying Qiu | 2 | 0 | 0.34 |
Yu Hong | 3 | 246 | 35.44 |
Yadong Chen | 4 | 0 | 0.34 |
Jianming Yao | 5 | 14 | 2.07 |
Qiaoming Zhu | 6 | 558 | 76.34 |
Guodong Zhou | 7 | 2528 | 192.09 |