The best of both worlds: a graph-based completion model for transition-based parsers (original) (raw)

Comparing advanced graph-based and transition-based dependency parsers

In this paper, we compare a higher order graph-based parser and a transitionbased parser with beam search. These parsers provide a higher accuracy than a second order MST parser and a deterministic transition-based parser. We apply and compare the output on languages, which have not been in the research focus of Shared Tasks. The parser are implemented in a uniform framework. The transitionbased parser was newley implemented and we revised the graph-based parser. The graph-based parser has to our knowlege the highest published scores for French and Czech with 90.40 and 81.43 labeled accuracy score.

A Transition-Based Dependency Parser Using a Dynamic Parsing Strategy

2013

We present a novel transition-based, greedy dependency parser which implements a flexible mix of bottom-up and top-down strategies. The new strategy allows the parser to postpone difficult decisions until the relevant information becomes available. The novel parser has a 12% error reduction in unlabeled attachment score over an arc-eager parser, with a slow-down factor of 2.8.

Dependency parsing with undirected graphs

2012

Abstract We introduce a new approach to transitionbased dependency parsing in which the parser does not directly construct a dependency structure, but rather an undirected graph, which is then converted into a directed dependency tree in a post-processing step. This alleviates error propagation, since undirected parsers do not need to observe the single-head constraint.

A transition-based parser for 2-planar dependency structures

… of the 48th Annual Meeting of the …, 2010

Finding a class of structures that is rich enough for adequate linguistic representation yet restricted enough for efficient computational processing is an important problem for dependency parsing. In this paper, we present a transition system for 2-planar dependency trees -trees that can be decomposed into at most two planar graphs -and show that it can be used to implement a classifier-based parser that runs in linear time and outperforms a stateof-the-art transition-based parser on four data sets from the CoNLL-X shared task. In addition, we present an efficient method for determining whether an arbitrary tree is 2-planar and show that 99% or more of the trees in existing treebanks are 2-planar.

Dependency Language Models for Transition-based Dependency Parsing

ArXiv, 2017

In this paper, we present an approach to improve the accuracy of a strong transition-based dependency parser by exploiting dependency language models that are extracted from a large parsed corpus. We integrated a small number of features based on the dependency language models into the parser. To demonstrate the effectiveness of the proposed approach, we evaluate our parser on standard English and Chinese data where the base parser could achieve competitive accuracy scores. Our enhanced parser achieved state-of-the-art accuracy on Chinese data and competitive results on English data. We gained a large absolute improvement of one point (UAS) on Chinese and 0.5 points for English.

Training Parsers on Partial Trees: A Cross-language Comparison

Language Resources and Evaluation, 2010

We present a study that compares data-driven dependency parsers obtained by means of annotation projection between language pairs of varying structural similarity. We show how the partial dependency trees projected from English to Dutch, Italian and German can be exploited to train parsers for the target languages. We evaluate the parsers against manual gold standard annotations and find that the projected parsers substantially outperform our heuristic baseline by 9-25% UAS, which corresponds to a 21-43% reduction in error rate. A comparative error analysis focuses on how the projected target language parsers handle subjects, which is especially interesting for Italian as an instance of a pro-drop language. For Dutch, we further present experiments with German as an alternative source language. In both source languages, we contrast standard baseline parsers with parsers that are enhanced with the predictions from large-scale LFG grammars through a technique of parser stacking, and show that improvements of the source language parser can directly lead to similar improvements of the projected target language parser.

CUNI: Feature Selection and Error Analysis of a Transition-Based Parser

2012

We describe the parsing system used at the Charles University (CUNI) for the Hindi Parsing Shared Task 2012. We used the publicly available Malt Parser, which is highly configurable. A substantial part of the paper describes the configuration that we selected. The parser performs reasonably well in identifying the head nodes. The main weakness is in labeling the dependency relations. We identify the most prominent error types, which should help to improve the parsing accuracy in future. Title and Abstract in Czech CUNI: Výběr rysů a analýza chyb parseru založeneho na přechodech Popisujeme system pro syntaktickou analýzu použitý na Univerzitě Karlově (CUNI) pro Hindi Parsing Shared Task 2012. Použili jsme veřejně dostupný nastroj Malt Parser, který poskytuje mnoho možnosti konfigurace. Podstatna cast clanku se zabýva pravě konfiguraci, kterou jsme zvolili. Parser dosahuje dobre uspěsnosti při identifikaci rodicovských uzlů. Jeho hlavni slabinou je znackovani zavislostnich vztahů. Pop...

Sentence-Level Instance-Weighting for Graph-Based and Transition-Based Dependency Parsing

Conference on Parsing Technologies, 2011

Instance-weighting has been shown to be effective in statistical machine translation (Foster et al., 2010), as well as crosslanguage adaptation of dependency parsers (Søgaard, 2011). This paper presents new methods to do instance-weighting in stateof-the-art dependency parsers. The methods are evaluated on Danish and English data with consistent improvements over unadapted baselines.

Inspecting the structural biases of dependency parsing algorithms

We propose the notion of a structural bias inherent in a parsing system with respect to the language it is aiming to parse. This structural bias characterizes the behaviour of a parsing system in terms of structures it tends to under- and over- produce. We propose a Boosting-based method for uncovering some of the structural bias inherent in parsing systems. We then apply our method to four English dependency parsers (an Arc-Eager and Arc-Standard transition-based parsers, and first- and second-order graph-based parsers). We show that all four parsers are biased with respect to the kind of annotation they are trained to parse. We present a detailed analysis of the biases that highlights specific differences and commonalities between the parsing systems, and improves our understanding of their strengths and weaknesses.