Current Directions in English and Arabic PropBank

Bonial, Claire; Conger, Kathryn; Hwang, Jena D.; Mansouri, Aous; Aseri, Yahya; Bonn, Julia; O’Gorman, Timothy; Palmer, Martha

doi:10.1007/978-94-024-0881-2_27

Cited by 25 publications

(26 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The syntactic dependencies are obtained from CCProcessed dependencies output by applying Stanford dependencies converter ( 16 ) on a parse tree obtained by Bllip parser ( 17 ). The other type of dependencies is the numbered arguments, whose idea is based on the guidelines of PropBank ( 14 ). For the PPI detection task, we use only arg0 and arg1 in EDG.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

BioC-compatible full-text passage detection for protein–protein interactions using extended dependency graph

Peng

Arighi

et al. 2016

Database

View full text Add to dashboard Cite

There has been a large growth in the number of biomedical publications that report experimental results. Many of these results concern detection of protein–protein interactions (PPI). In BioCreative V, we participated in the BioC task and developed a PPI system to detect text passages with PPIs in the full-text articles. By adopting the BioC format, the output of the system can be seamlessly added to the biocuration pipeline with little effort required for the system integration. A distinctive feature of our PPI system is that it utilizes extended dependency graph, an intermediate level of representation that attempts to abstract away syntactic variations in text. As a result, we are able to use only a limited set of rules to extract PPI pairs in the sentences, and additional rules to detect additional passages for PPI pairs. For evaluation, we used the 95 articles that were provided for the BioC annotation task. We retrieved the unique PPIs from the BioGRID database for these articles and show that our system achieves a recall of 83.5%. In order to evaluate the detection of passages with PPIs, we further annotated Abstract and Results sections of 20 documents from the dataset and show that an f-value of 80.5% was obtained. To evaluate the generalizability of the system, we also conducted experiments on AIMed, a well-known PPI corpus. We achieved an f-value of 76.1% for sentence detection and an f-value of 64.7% for unique PPI detection.Database URL: http://proteininformationresource.org/iprolink/corpora

show abstract

Section: Methodsmentioning

confidence: 99%

“…Third, to test the generalizability of our system as well as the precision of PPI detection, we evaluated the system on AIMed ( 13 , 14 ), which is a widely-used PPI corpus. Since this corpus contains annotations with individual PPI mentions, we can use it for both PPI sentence detection and PPI pair detection.…”

Section: Introductionmentioning

confidence: 99%

BioC-compatible full-text passage detection for protein–protein interactions using extended dependency graph

Peng

Arighi

et al. 2016

Database

View full text Add to dashboard Cite

show abstract

“…Offer-01 3 Arg0: entity offering Arg1: commodity Arg2: price Previous versions of PropBank maintained separate rolesets for distinct parts of speech (e.g. the verb offer, and the related nouns offer/offering) but, as part of this body of research, these have been combined to provide parallel annotations of related usages (Bonial et al 2014(Bonial et al , 2017. For example: 5 Prior to this research (see also Hwang et al 2010), PropBank 1.0 had no special guidelines for LVCs.…”

Section: Development Of the Propbank Lvc Annotation Schemamentioning

confidence: 99%

Choosing an event description: What a PropBank study reveals about the contrast between light verb constructions and counterpart synthetic verbs

Bonial

Pollard

2020

J. Ling.

Self Cite

View full text Add to dashboard Cite

Light verb constructions (LVCs) in English and Romance languages are somewhat unique crosslinguistically because LVCs in these languages tend to have semantically similar synthetic verb counterparts (Zarco 1999): e.g. make an appearance and appear. This runs contrary to assumptions in linguistic theories that two competing forms are rarely maintained in a language unless they serve distinct purposes (e.g. Grice 1975). Why do English LVCs exist alongside counterpart synthetic verbs, especially given that synthetic verbs are arguably the more efficient form (Zipf 1949)? It has been proposed that LVCs serve an aspectual function (Prince 1972, Live 1973, Wierzbicka 1982, Tanabe 1999, Butt & Geuder 2001), as there are telic LVC counterparts (e.g. have a thought) of atelic verbs (e.g. think). This proposal has been difficult to evaluate without a large-scale resource providing a markup of both LVCs and counterpart verbs. Addressing this gap in resources, the present research describes the development of guidelines for LVC annotation in the English PropBank (Bonial & Palmer 2015). The focus of this article is the subsequent analysis of these annotations, aimed at uncovering corpus evidence of what contexts call for the use of an LVC over a synthetic verb. The corpus study shows that the general function of LVCs is not an aspectual one and provides distributional evidence that the ease and variety with which LVCs can be modified is the general motivating factor for the use of an LVC.

show abstract

“…Annotated selections from various genres (including newswire, discussion forum, other web logs, and television transcripts) are available, for a total of 39,260 sentences. This release uses the PropBank Unification frame files (Bonial et al, 2014;Bonial et al, 2016). To generate automatic dependency parses for all DEFT AMR Release data, we use ClearNLP (Choi and Mccallum, 2013) to produce dependency parses.…”

Section: Experimental Datamentioning

confidence: 99%

Unsupervised AMR-Dependency Parse Alignment

Chen

Palmer

2017

Proceedings of the 15th Conference of the European Chapter of The Association for Computational Linguistics: Volume 1

Self Cite

View full text Add to dashboard Cite

In this paper, we introduce an Abstract Meaning Representation (AMR) to Dependency Parse aligner. Alignment is a preliminary step for AMR parsing, and our aligner improves current AMR parser performance. Our aligner involves several different features, including named entity tags and semantic role labels, and uses Expectation-Maximization training. Results show that our aligner reaches an 87.1% F-Score score with the experimental data, and enhances AMR parsing.

show abstract

Current Directions in English and Arabic PropBank

Cited by 25 publications

References 13 publications

BioC-compatible full-text passage detection for protein–protein interactions using extended dependency graph

BioC-compatible full-text passage detection for protein–protein interactions using extended dependency graph

Choosing an event description: What a PropBank study reveals about the contrast between light verb constructions and counterpart synthetic verbs

Unsupervised AMR-Dependency Parse Alignment

Contact Info

Product

Resources

About