Extracting Relational Facts by an End-to-End Neural Model with Copy Mechanism | |
Xiangrong Zeng1,2; Daojian Zeng3; Shizhu He2; Kang Liu1,2; Jun Zhao1,2 | |
2018 | |
会议日期 | Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (ACL2018) |
会议地点 | 澳大利亚 |
英文摘要 | The relational facts in sentences are often complicated. Different relational triplets may have overlaps in a sentence. We divided the sentences into three types according to triplet overlap degree, including Normal, EntityPairOverlap and SingleEntiyOverlap. Existing methods mainly focus on \emph{Normal} class and fail to extract relational triplets precisely. In this paper, we propose an end-to-end model based on sequence-to-sequence learning with copy mechanism, which can jointly extract relational facts from sentences of any of these classes. We adopt two different strategies in decoding process: employing only one united decoder or applying multiple separated decoders. We test our models in two public datasets and our model outperform the baseline method significantly. |
产权排序 | 1 |
内容类型 | 会议论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/22073] |
专题 | 自动化研究所_模式识别国家重点实验室_自然语言处理团队 自然语言处理团队 |
作者单位 | 1.中国科学院大学 2.中科院自动化所 3.长沙理工大学 |
推荐引用方式 GB/T 7714 | Xiangrong Zeng,Daojian Zeng,Shizhu He,et al. Extracting Relational Facts by an End-to-End Neural Model with Copy Mechanism[C]. 见:. 澳大利亚. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (ACL2018). |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论