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 1808.04850cs.CL
收藏

神经元成分分析的两种局部模型

Two Local Models for Neural Constituent Parsing

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

Zhiyang Teng, Yue Zhang

摘要:

非局部特征已经被句法分析器用于捕获子输出结构之间的依赖关系。这些特性是最先进的统计解析器成功的关键。然而,随着深度学习的兴起,已经表明,由于包含全局语法信息的密集神经输入表示的力量,局部输出决策可以提供极具竞争力的准确性。我们研究了两个概念上简单的局部神经元模型的成分分析,使当地的决定,成分跨度和CFG规则,分别。与之前的研究结果一致,我们的最佳模型给出了极具竞争力的结果,在PTB上实现了92.4%的标记括号F1评分,在CTB 5.1上实现了87.3%。

英文摘要:

Non-local features have been exploited by syntactic parsers for capturing dependencies between sub output structures. Such features have been a key to the success of state-of-the-art statistical parsers. With the rise of deep learning, however, it has been shown that local output decisions can give highly competitive accuracies, thanks to the power of dense neural input representations that embody global syntactic information. We investigate two conceptually simple local neural models for constituent parsing, which make local decisions to constituent spans and CFG rules, respectively. Consistent with previous findings along the line, our best model gives highly competitive results, achieving the labeled bracketing F1 scores of 92.4% on PTB and 87.3% on CTB 5.1.

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。

↑