Faster shift-reduce constituent parsing with a non-binary, bottom-up strategy

DOI10.1016/J.ARTINT.2019.07.006zbMATH Open1478.68415arXiv1804.07961OpenAlexW2799211465WikidataQ127480147 ScholiaQ127480147MaRDI QIDQ2321341FDOQ2321341

Daniel Fernández-González, C. Gómez-Rodríguez

Publication date: 28 August 2019

Published in: Artificial Intelligence (Search for Journal in Brave)

Abstract: An increasingly wide range of artificial intelligence applications rely on syntactic information to process and extract meaning from natural language text or speech, with constituent trees being one of the most widely used syntactic formalisms. To produce these phrase-structure representations from sentences in natural language, shift-reduce constituent parsers have become one of the most efficient approaches. Increasing their accuracy and speed is still one of the main objectives pursued by the research community so that artificial intelligence applications that make use of parsing outputs, such as machine translation or voice assistant services, can improve their performance. With this goal in mind, we propose in this article a novel non-binary shift-reduce algorithm for constituent parsing. Our parser follows a classical bottom-up strategy but, unlike others, it straightforwardly creates non-binary branchings with just one Reduce transition, instead of requiring prior binarization or a sequence of binary transitions, allowing its direct application to any language without the need of further resources such as percolation tables. As a result, it uses fewer transitions per sentence than existing transition-based constituent parsers, becoming the fastest such system and, as a consequence, speeding up downstream applications. Using static oracle training and greedy search, the accuracy of this novel approach is on par with state-of-the-art transition-based constituent parsers and outperforms all top-down and bottom-up greedy shift-reduce systems on the Wall Street Journal section from the English Penn Treebank and the Penn Chinese Treebank. Additionally, we develop a dynamic oracle for training the proposed transition-based algorithm, achieving further improvements in both benchmarks and obtaining the best accuracy to date on the Penn Chinese Treebank among greedy shift-reduce parsers.

Full work available at URL: https://arxiv.org/abs/1804.07961

zbMATH Keywords

automataparsingcomputational linguisticsnatural language processingconstituent parsing

Mathematics Subject Classification ID

Natural language processing (68T50)

Cites Work

• Three models for the description of language
• Finding the smallest binarization of a CFG is NP-hard
• Title not available (Why is that?)

Cited In (1)

• Strahler number of natural language sentences in comparison with random trees

Uses Software

• LSTM
• Stanford Tagger
• Penn Treebank
• SQuAD

   Recommendations

• Even faster generalized LR parsing 👍 👎
• Title not available (Why is that?) 👍 👎
• Shift-Reduce Parsers for Transition Networks 👍 👎
• Predictive Shift-Reduce Parsing for Hyperedge Replacement Grammars 👍 👎
• Extending predictive shift-reduce parsing to contextual hyperedge replacement grammars 👍 👎

This page was built for publication: Faster shift-reduce constituent parsing with a non-binary, bottom-up strategy

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2321341)