Partial parse selection for robust deep processing

Zhang, Yi; Kordoni, Valia; Fitzgerald, Erin

doi:10.3115/1608912.1608932

Cited by 5 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this experiment, we used English Resource Grammar (ERG, Flickinger (2000)), a precision HPSG grammar for English. For parsing, we used PET (Callmeier, 2001), an efficient HPSG parser, together with extended partial parsing module (Zhang et al, 2007) for maximum robustness. The grammar is handcrafted and the disambiguation models are trained on Redwoods treebanks.…”

Section: Featuresmentioning

confidence: 99%

Hybrid learning of dependency structures from heterogeneous linguistic resources

Zhang¹,

Wang²,

Uszkoreit³

2008

Proceedings of the Twelfth Conference on Computational Natural Language Learning - CoNLL '08

Self Cite

View full text Add to dashboard Cite

In this paper we present our syntactic and semantic dependency parsing system participated in both closed and open competitions of the CoNLL 2008 Shared Task. By combining the outcome of two state-ofthe-art syntactic dependency parsers, we achieved high accuracy in syntactic dependencies (87.32%). With MRSes from grammar-based HPSG parsers, we achieved significant performance improvement on semantic role labeling (from 71.31% to 71.89%), especially in the out-domain evaluation (from 60.16% to 62.11%).

show abstract

Section: Featuresmentioning

confidence: 99%

Hybrid learning of dependency structures from heterogeneous linguistic resources

Zhang¹,

Wang²,

Uszkoreit³

2008

Proceedings of the Twelfth Conference on Computational Natural Language Learning - CoNLL '08

Self Cite

View full text Add to dashboard Cite

show abstract

“…Since these rules are sometimes too restrictive and only produce a partial chart, a second stage with a permissive pcfg backbone predicts a full-spanning pseudo-derivation, and the fragmented mrs structures are extracted from these partial analyses. (Zhang et al, 2007). The fragmented mrs does not have full feature structures for all constituents because of rule conflict, otherwise it could have been parsed in the first stage.…”

Section: Parsing With Petmentioning

confidence: 99%

Semantics-based Question Generation and Implementation

Xiao

Bouma

Zhang³

2012

dad

Self Cite

View full text Add to dashboard Cite

This paper presents a question generation system based on the approach of semantic rewriting. The state-of-the-art deep linguistic parsing and generation tools are employed to convert (back and forth) between the natural language sentences and their meaning representations in the form of Minimal Recursion Semantics (MRS). By carefully operating on the semantic structures, we show a principled way of generating questions without ad-hoc manipulation of the syntactic structures. Based on the (partial) understanding of the sentence meaning, the system generates questions which are semantically grounded and purposeful. And with the support of deep linguistic grammars, the grammaticality of the generation results is warranted. Further, with a specialized ranking model, the linguistic realizations from the general purpose generation model are further refined for our the question generation task. The evaluation results from QGSTEC2010 show promising prospects of the proposed approach.

show abstract

“…Several methods were proposed for robust analysis. Partial parse is defined in [22] as the set of intermediate passive edges of the chart whose spans are not overlapping and cover the entire input sequence. For selecting the best sequence of partial 6Fernando López-Colino, Javier Tejedor, Jordi Porta and José Colás parses, they apply the shortest-path algorithm to maximal projections, i.e., complete phrases.…”

Section: Fig 3 Transferred Dependency Tree (Left) and Its Corresponmentioning

confidence: 99%

Integration of a Spanish-to-LSE Machine Translation System into an e-learning Platform

López-Colino¹,

Tejedor²,

Porta³

et al. 2011

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Esta es la versión de autor de la comunicación de congreso publicada en: This is an author produced version of a paper published in: Abstract. This paper presents the first results of the integration of a Spanish-to-LSE Machine Translation (MT) system into an e-learning platform. Most elearning platforms provide speech-based contents, which makes them inaccessible to the Deaf. To solve this issue, we have developed a MT system that translates Spanish speech-based contents into LSE. To test our MT system, we have integrated it into an e-learning tool. The elearning tool sends the audio to our platform. The platform sends back the subtitles and a video stream with the signed translation to the e-learning tool. Preliminary results, evaluating the sign language synthesis module, show an isolated sign recognition accuracy of 97%. The sentence recognition accuracy was of 93%.

show abstract

Partial parse selection for robust deep processing

Cited by 5 publications

References 17 publications

Hybrid learning of dependency structures from heterogeneous linguistic resources

Hybrid learning of dependency structures from heterogeneous linguistic resources

Semantics-based Question Generation and Implementation

Integration of a Spanish-to-LSE Machine Translation System into an e-learning Platform

Contact Info

Product

Resources

About