arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1807.01396cs.CL
收藏

更简单但更精确的语义依存句法分析

Simpler but More Accurate Semantic Dependency Parsing

Stanford University(斯坦福大学)

Timothy Dozat, Christopher D. Manning

摘要:

虽然句法依赖性标注集中在句子的表面或功能结构上,但语义依赖性标注旨在使用图形结构表示来捕获与句子的含义更密切相关的词间关系。我们扩展了Dozat和Manning(2017)基于LSTM的语法解析器,以训练并生成这些图结构。由此产生的系统本身实现了最先进的性能,比之前的更复杂的最先进的系统高出0.6%,标记为F1。添加语言上更丰富的输入表示将边际推得更高,使我们能够以1.9%的F1击败它。

英文摘要:

While syntactic dependency annotations concentrate on the surface or functional structure of a sentence, semantic dependency annotations aim to capture between-word relationships that are more closely related to the meaning of a sentence, using graph-structured representations. We extend the LSTM-based syntactic parser of Dozat and Manning (2017) to train on and generate these graph structures. The resulting system on its own achieves state-of-the-art performance, beating the previous, substantially more complex state-of-the-art system by 0.6% labeled F1. Adding linguistically richer input representations pushes the margin even higher, allowing us to beat it by 1.9% labeled F1.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑