arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1612.00567cs.CL
收藏

基于神经元前瞻特征的Shift-Reduce成分分析

Shift-Reduce Constituent Parsing with Neural Lookahead Features

Singapore University of Technology and Design(新加坡科技与设计大学)

Jiangming Liu, Yue Zhang

摘要:

基于转换的模型可以快速准确地进行成分分析。与基于图表的模型相比,它们通过从跨越非局部成分的解析器堆栈中提取历史信息来利用更丰富的特征。另一方面,在增量解析期间,不利用当前单词右侧的成分信息,这是移位归约解析的相对弱点。为了解决这一限制,我们利用快速神经模型来提取前瞻特征。特别是,我们构建了一个双向LSTM模型,它利用完整的句子信息来预测每个单词开始和结束的成分的层次结构。然后将结果作为前瞻特征传递给基于强转换的成分解析器。与基线相比,所得到的解析器在WSJ中给出了1.3%的绝对改进,在CTB中给出了2.3%的绝对改进,给出了全监督解析的最高准确率。

英文摘要:

Transition-based models can be fast and accurate for constituent parsing. Compared with chart-based models, they leverage richer features by extracting history information from a parser stack, which spans over non-local constituents. On the other hand, during incremental parsing, constituent information on the right hand side of the current word is not utilized, which is a relative weakness of shift-reduce parsing. To address this limitation, we leverage a fast neural model to extract lookahead features. In particular, we build a bidirectional LSTM model, which leverages the full sentence information to predict the hierarchy of constituents that each word starts and ends. The results are then passed to a strong transition-based constituent parser as lookahead features. The resulting parser gives 1.3% absolute improvement in WSJ and 2.3% in CTB compared to the baseline, given the highest reported accuracies for fully-supervised parsing.

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。

↑