Multiword Expression Translation Using Generative Dependency Grammar

Diaconescu, Ştefan

doi:10.1007/978-3-540-30228-5_22

Cited by 24 publications

(6 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although these lexicons suffer from insufficient formalization , they could be successfully applied to parsing after ad hoc conversion to particular grammar formalisms. On the other hand, some approaches accommodate some types of MWEs directly in the lexicons of computational grammars within particular grammatical frameworks: head-driven phrase structure grammar Villavicencio et al, 2004;Bond et al, 2015;Herzig Sheinfux et al, 2015), lexical functional grammar (Attia, 2006;Dyvik et al, 2019), tree-adjoining grammar Schabes, 1989, 1996;Vaidya et al, 2014;Lichte and Kallmeyer, 2016), and dependency grammar (Diaconescu, 2004).…”

Section: Lexicons Of Mwesmentioning

confidence: 99%

Without lexicons, multiword expression identification will never fly: A position statement

Savary¹,

Cordeiro²,

Ramisch³

2019

Proceedings of the Joint Workshop on Multiword Expressions and WordNet (MWE-WN 2019)

View full text Add to dashboard Cite

Because most multiword expressions (MWEs), especially verbal ones, are semantically non-compositional, their automatic identification in running text is a prerequisite for semantically-oriented downstream applications. However, recent developments, driven notably by the PARSEME shared task on automatic identification of verbal MWEs, show that this task is harder than related tasks, despite recent contributions both in multilingual corpus annotation and in computational models. In this paper, we analyse possible reasons for this state of affairs. They lie in the nature of the MWE phenomenon, as well as in its distributional properties. We also offer a comparative analysis of the state-of-the-art systems, which exhibit particularly strong sensitivity to unseen data. On this basis, we claim that, in order to make strong headway in MWE identification, the community should bend its mind into coupling identification of MWEs with their discovery, via syntactic MWE lexicons. Such lexicons need not necessarily achieve a linguistically complete modelling of MWEs' behavior, but they should provide minimal morphosyntactic information to cover some potential uses, so as to complement existing MWE-annotated corpora. We define requirements for such a minimal NLP-oriented lexicon, and we propose a roadmap for the MWE community driven by these requirements.

show abstract

Section: Lexicons Of Mwesmentioning

confidence: 99%

Without lexicons, multiword expression identification will never fly: A position statement

Savary¹,

Cordeiro²,

Ramisch³

2019

Proceedings of the Joint Workshop on Multiword Expressions and WordNet (MWE-WN 2019)

View full text Add to dashboard Cite

show abstract

“…începerea formalizării morfologiei limbii române; 3. finalizarea formalizării morfologiei limbii române, formalizarea silabației, precum și a accentuării cuvintelor; 4. realizarea Dicționarului morfologic român; 5. formalizarea fonologiei și adăugarea de reguli și atribute fonologice la resursele lingvistice existente; 6. extinderea Dicţionarului morfologic român, precum și "analiza sintactico-semantică a textelor româneşti, cu aplicaţii dintre cele mai diverse: corectoare gramaticale, sisteme de asistare a învăţării limbii române (atât de către vorbitorii nativi, cât şi de către cei străini), sisteme de adnotare a corpusurilor, sisteme de traducere automată etc." (Tămâianu-Morita et al 2006-2007.…”

Section: Descriereunclassified

Un inventar al lexicoanelor (digitale) pentru limba română. Perioada 1973–2022

Teodorescu

2023

ALIL

View full text Add to dashboard Cite

This article aims to present the lexicons (or resources that are similar to lexicons), which were elaborated in Romania during the period 1973–2022, in all media forms, from the written to the digital. The term lexicon is hereby understood as it is used within the field of generative grammar. For each project I considered the following pieces of information: description, objectives, team, funding, input data, work methodology, data model and, respectively, the results. As to the data model, I have made terminological conversions, due to the fact that some projects, especially those realised by researchers with an Information Technology background, use a terminology that is more specific to Information Technology field than to linguistics or philology.

show abstract

“…Different solutions have been proposed in order to guarantee proper handling of MWEs in an MT process. Diaconescu (2004) stresses the difficulties of MWE processing in MT and proposes a method based on Generative Dependency Grammars with features. Lambert & Banchs (2006) suggest a strategy for identifying and using MWEs in SMT, based on grouping bilingual MWEs before performing statistical alignment.…”

Section: Related Workmentioning

confidence: 99%

ALCIDE: An online platform for the analysis of language and content in a digital environment

Moretti¹,

Tonelli²,

Menini³

2014

Proceedings of the First Italian Conference on Computational Linguistics CLiC-it 2014 and of the Fourth International Workshop

View full text Add to dashboard Cite

We are glad to introduce CLiC-it 2014 (http://clic.humnet.unipi.it), the first edition of the Italian Conference on Computational Linguistics, a new event aiming to establish a reference forum for research on Computational Linguistics of the Italian community. CLiC-it covers all aspects of automatic language understanding, both written and spoken, and targets state-of-art theoretical results, experimental methodologies, technologies, as well as application perspectives, which may contribute to advance the field.CLiC-it 2014 is held in Pisa on December 9-10 2014, and it is co-located with EVALITA-2014 (http://www.evalita.it), the fourth edition of the evaluation campaign of Natural Language Processing and Speech tools for Italian and with the XIII Symposium on Artificial Intelligence (Pisa, 10-12 December 2014, http://aiia2014.di.unipi.it/).Pisa is a special place in the history of Italian Computational Linguistics. Here, Padre Roberto Busa carried out his pioneering research on automatic text processing in the late '60s with Antonio Zampolli, who then founded the Istituto di Linguistica Computazionale of CNR in Pisa, the first research center thoroughly devoted to Computational Linguistics and Natural Language Processing. The University of Pisa also hosted the first professorship in Computational Linguistics held by Antonio Zampolli until his death in 2003.It is therefore highly symbolic that the Italian community on Computational Linguistics gathers for the first time in Pisa, there where its roots lie. Italian Computational Linguistics has come a long way. Research groups and centers are now spread nationwide and play an active role on the international scene. The large number of researchers that have decided to present their work to CLiC-it is the best proof of the maturity of our community, strongly committed to shape the future of Computational Linguistics.The spirit of CLiC-it is inclusive. In the conviction that the complexity of language phenomena needs cross-disciplinary competences, CLiC-it intends to bring together researchers of related disciplines such as

show abstract

Multiword Expression Translation Using Generative Dependency Grammar

Cited by 24 publications

References 4 publications

Without lexicons, multiword expression identification will never fly: A position statement

Without lexicons, multiword expression identification will never fly: A position statement

Un inventar al lexicoanelor (digitale) pentru limba română. Perioada 1973–2022

ALCIDE: An online platform for the analysis of language and content in a digital environment

Contact Info

Product

Resources

About