Inducción automática de una taxonomía multilingüe de marcadores discursivos: primeros resultados en castellano, inglés, francés, alemán y catalán

Rogelio Nazar

doi:10.26342/2021-67-11

Inducción automática de una taxonomía multilingüe de marcadores discursivos: primeros resultados en castellano, inglés, francés, alemán y catalán

Translated title of the contribution: Automatic induction of a multilingual taxonomy of discourse markers: first results in Spanish, English, French, German and Catalan

Rogelio Nazar

INSTITUTE OF LITERATURE AND LANGUAGE SCIENCES

Research output: Contribution to journal › Article › peer-review

1 Scopus citations

Abstract

This paper presents a methodological proposal por the automatic induction of a multilingual taxonomy of discourse markers which, in the case of English, correspond to units such as however, therefore, by the way, etc. First, a method is proposed to separate such units from the rest of the vocabulary using a measure of information, followed by a method to group them using a parallel corpus. Finally, this categorization is used as the basis for the extraction and classification of new units. Apart from the method, the first results are described, which consist of a database that currently surpasses 2600 units.

Translated title of the contribution	Automatic induction of a multilingual taxonomy of discourse markers: first results in Spanish, English, French, German and Catalan
Original language	Spanish
Pages (from-to)	127-138
Number of pages	12
Journal	Procesamiento de Lenguaje Natural
Volume	67
DOIs	https://doi.org/10.26342/2021-67-11
State	Published - Sep 2021

Access to Document

10.26342/2021-67-11

Cite this

@article{6314e8618f994f1e894c5c7d9b2b25b2,

title = "Inducci{\'o}n autom{\'a}tica de una taxonom{\'i}a multiling{\"u}e de marcadores discursivos: primeros resultados en castellano, ingl{\'e}s, franc{\'e}s, alem{\'a}n y catal{\'a}n",

abstract = "This paper presents a methodological proposal por the automatic induction of a multilingual taxonomy of discourse markers which, in the case of English, correspond to units such as however, therefore, by the way, etc. First, a method is proposed to separate such units from the rest of the vocabulary using a measure of information, followed by a method to group them using a parallel corpus. Finally, this categorization is used as the basis for the extraction and classification of new units. Apart from the method, the first results are described, which consist of a database that currently surpasses 2600 units.",

keywords = "computational lexicography, discourse markers, discurse particles, taxonomy induction",

author = "Rogelio Nazar",

note = "Publisher Copyright: {\textcopyright} 2021 Sociedad Espa{\~n}ola para el Procesamiento del Lenguaje Natural",

year = "2021",

month = sep,

doi = "10.26342/2021-67-11",

language = "Spanish",

volume = "67",

pages = "127--138",

journal = "Procesamiento de Lenguaje Natural",

issn = "1135-5948",

publisher = "Sociedad Espanola para el Procesamiento del Lenguaje Natural",

}

TY - JOUR

T1 - Inducción automática de una taxonomía multilingüe de marcadores discursivos

T2 - primeros resultados en castellano, inglés, francés, alemán y catalán

AU - Nazar, Rogelio

PY - 2021/9

Y1 - 2021/9

N2 - This paper presents a methodological proposal por the automatic induction of a multilingual taxonomy of discourse markers which, in the case of English, correspond to units such as however, therefore, by the way, etc. First, a method is proposed to separate such units from the rest of the vocabulary using a measure of information, followed by a method to group them using a parallel corpus. Finally, this categorization is used as the basis for the extraction and classification of new units. Apart from the method, the first results are described, which consist of a database that currently surpasses 2600 units.

AB - This paper presents a methodological proposal por the automatic induction of a multilingual taxonomy of discourse markers which, in the case of English, correspond to units such as however, therefore, by the way, etc. First, a method is proposed to separate such units from the rest of the vocabulary using a measure of information, followed by a method to group them using a parallel corpus. Finally, this categorization is used as the basis for the extraction and classification of new units. Apart from the method, the first results are described, which consist of a database that currently surpasses 2600 units.

KW - computational lexicography

KW - discourse markers

KW - discurse particles

KW - taxonomy induction

UR - http://www.scopus.com/inward/record.url?scp=85127556705&partnerID=8YFLogxK

U2 - 10.26342/2021-67-11

DO - 10.26342/2021-67-11

M3 - Article

AN - SCOPUS:85127556705

SN - 1135-5948

VL - 67

SP - 127

EP - 138

JO - Procesamiento de Lenguaje Natural

JF - Procesamiento de Lenguaje Natural

ER -

Inducción automática de una taxonomía multilingüe de marcadores discursivos: primeros resultados en castellano, inglés, francés, alemán y catalán

Abstract

Access to Document

Other files and links

Fingerprint

Cite this