Linguistically-driven Selection of Difficult-to-Parse Dependency Structures

The paper illustrates a novel methodology meeting a twofold goal, namely quantifying the reliability of automatically generated dependency relations without using gold data on the one hand, and identifying which are the linguistic constructions negatively affecting the parser performance on the othe...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Chiara Alzetta, Felice Dell’Orletta, Simonetta Montemagni, Giulia Venturi
Formato: article
Lenguaje:EN
Publicado: Accademia University Press 2020
Materias:
H
Acceso en línea:https://doaj.org/article/d0572ebc690a4943b27521dfcad6c226
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:d0572ebc690a4943b27521dfcad6c226
record_format dspace
spelling oai:doaj.org-article:d0572ebc690a4943b27521dfcad6c2262021-12-02T09:52:35ZLinguistically-driven Selection of Difficult-to-Parse Dependency Structures2499-455310.4000/ijcol.719https://doaj.org/article/d0572ebc690a4943b27521dfcad6c2262020-12-01T00:00:00Zhttp://journals.openedition.org/ijcol/719https://doaj.org/toc/2499-4553The paper illustrates a novel methodology meeting a twofold goal, namely quantifying the reliability of automatically generated dependency relations without using gold data on the one hand, and identifying which are the linguistic constructions negatively affecting the parser performance on the other hand. These represent objectives typically investigated in different lines of research, with different methods and techniques. Our methodology, at the crossroads of these perspectives, allows not only to quantify the parsing reliability of individual dependency types, but also to identify and weight the contextual properties making relation instances more or less difficult to parse. The proposed methodology was tested in two different and complementary experiments, aimed at assessing the degree of parsing difficulty across (a) different dependency relation types, and (b) different instances of the same relation. The results show that the proposed methodology is able to identify difficult-to-parse dependency relations without relying on gold data and by taking into account a variety of intertwined linguistic factors. These findings pave the way to novel applications of the methodology, both in the direction of defining new evaluation metrics based purely on automatically parsed data and towards the automatic creation of challenge sets.Chiara AlzettaFelice Dell’OrlettaSimonetta MontemagniGiulia VenturiAccademia University PressarticleSocial SciencesHComputational linguistics. Natural language processingP98-98.5ENIJCoL, Vol 6, Iss 2, Pp 37-60 (2020)
institution DOAJ
collection DOAJ
language EN
topic Social Sciences
H
Computational linguistics. Natural language processing
P98-98.5
spellingShingle Social Sciences
H
Computational linguistics. Natural language processing
P98-98.5
Chiara Alzetta
Felice Dell’Orletta
Simonetta Montemagni
Giulia Venturi
Linguistically-driven Selection of Difficult-to-Parse Dependency Structures
description The paper illustrates a novel methodology meeting a twofold goal, namely quantifying the reliability of automatically generated dependency relations without using gold data on the one hand, and identifying which are the linguistic constructions negatively affecting the parser performance on the other hand. These represent objectives typically investigated in different lines of research, with different methods and techniques. Our methodology, at the crossroads of these perspectives, allows not only to quantify the parsing reliability of individual dependency types, but also to identify and weight the contextual properties making relation instances more or less difficult to parse. The proposed methodology was tested in two different and complementary experiments, aimed at assessing the degree of parsing difficulty across (a) different dependency relation types, and (b) different instances of the same relation. The results show that the proposed methodology is able to identify difficult-to-parse dependency relations without relying on gold data and by taking into account a variety of intertwined linguistic factors. These findings pave the way to novel applications of the methodology, both in the direction of defining new evaluation metrics based purely on automatically parsed data and towards the automatic creation of challenge sets.
format article
author Chiara Alzetta
Felice Dell’Orletta
Simonetta Montemagni
Giulia Venturi
author_facet Chiara Alzetta
Felice Dell’Orletta
Simonetta Montemagni
Giulia Venturi
author_sort Chiara Alzetta
title Linguistically-driven Selection of Difficult-to-Parse Dependency Structures
title_short Linguistically-driven Selection of Difficult-to-Parse Dependency Structures
title_full Linguistically-driven Selection of Difficult-to-Parse Dependency Structures
title_fullStr Linguistically-driven Selection of Difficult-to-Parse Dependency Structures
title_full_unstemmed Linguistically-driven Selection of Difficult-to-Parse Dependency Structures
title_sort linguistically-driven selection of difficult-to-parse dependency structures
publisher Accademia University Press
publishDate 2020
url https://doaj.org/article/d0572ebc690a4943b27521dfcad6c226
work_keys_str_mv AT chiaraalzetta linguisticallydrivenselectionofdifficulttoparsedependencystructures
AT felicedellorletta linguisticallydrivenselectionofdifficulttoparsedependencystructures
AT simonettamontemagni linguisticallydrivenselectionofdifficulttoparsedependencystructures
AT giuliaventuri linguisticallydrivenselectionofdifficulttoparsedependencystructures
_version_ 1718397970740674560