Tarántula -> araña -> Animal: Asignación de hiperónimos de segundo nivel basada en métodos de similitud distribucional

Rogelio Nazar; Javier Obreque; Irene Renau

doi:10.26342/2020-64-3

Tarántula -> araña -> Animal: Asignación de hiperónimos de segundo nivel basada en métodos de similitud distribucional

Translated title of the contribution: Tarantula -> spider -> animal: Second level hypernymy discovery based on distributional similarity methods

Rogelio Nazar, Javier Obreque, Irene Renau

INSTITUTE OF LITERATURE AND LANGUAGE SCIENCES

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

Abstract

Automatic hypernymy discovery continues to present challenges for natural language processing. Polysemous nouns are linked to more than one hypernym and can therefore cause structural damage on a lexical taxonomy. For instance, the Spanish noun tarántula ('tarantula') is a hyponym of araña ('spider'), but this is also a polysemous noun, as it means 'chandelier' as well. It is thus necessary to determine the next hypernym in the chain, that is animal ('animal') or artefacto ('artifact'). In this paper we explore methods to solve this problem using a similarity measure that uses verb-noun co-occurrence as a predictor variable. Best results (84 % success) are obtained with a simple method that only measures co-occurrence, irrespective of any syntactic information.

Translated title of the contribution	Tarantula -> spider -> animal: Second level hypernymy discovery based on distributional similarity methods
Original language	Spanish
Pages (from-to)	29-36
Number of pages	8
Journal	Procesamiento de Lenguaje Natural
Volume	64
DOIs	https://doi.org/10.26342/2020-64-3
State	Published - Mar 2020

Access to Document

10.26342/2020-64-3

Cite this

@article{bc981afdcf024cbaa8a5a4882586e0bf,

title = "Tar{\'a}ntula -> ara{\~n}a -> Animal: Asignaci{\'o}n de hiper{\'o}nimos de segundo nivel basada en m{\'e}todos de similitud distribucional",

abstract = "Automatic hypernymy discovery continues to present challenges for natural language processing. Polysemous nouns are linked to more than one hypernym and can therefore cause structural damage on a lexical taxonomy. For instance, the Spanish noun tar{\'a}ntula ('tarantula') is a hyponym of ara{\~n}a ('spider'), but this is also a polysemous noun, as it means 'chandelier' as well. It is thus necessary to determine the next hypernym in the chain, that is animal ('animal') or artefacto ('artifact'). In this paper we explore methods to solve this problem using a similarity measure that uses verb-noun co-occurrence as a predictor variable. Best results (84 % success) are obtained with a simple method that only measures co-occurrence, irrespective of any syntactic information.",

keywords = "Distributional similarity, Hypernymy, Polysemy, Taxonomy",

author = "Rogelio Nazar and Javier Obreque and Irene Renau",

year = "2020",

month = mar,

doi = "10.26342/2020-64-3",

language = "Spanish",

volume = "64",

pages = "29--36",

journal = "Procesamiento de Lenguaje Natural",

issn = "1135-5948",

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

}

TY - JOUR

T1 - Tarántula -> araña -> Animal

T2 - Asignación de hiperónimos de segundo nivel basada en métodos de similitud distribucional

AU - Nazar, Rogelio

AU - Obreque, Javier

AU - Renau, Irene

PY - 2020/3

Y1 - 2020/3

N2 - Automatic hypernymy discovery continues to present challenges for natural language processing. Polysemous nouns are linked to more than one hypernym and can therefore cause structural damage on a lexical taxonomy. For instance, the Spanish noun tarántula ('tarantula') is a hyponym of araña ('spider'), but this is also a polysemous noun, as it means 'chandelier' as well. It is thus necessary to determine the next hypernym in the chain, that is animal ('animal') or artefacto ('artifact'). In this paper we explore methods to solve this problem using a similarity measure that uses verb-noun co-occurrence as a predictor variable. Best results (84 % success) are obtained with a simple method that only measures co-occurrence, irrespective of any syntactic information.

AB - Automatic hypernymy discovery continues to present challenges for natural language processing. Polysemous nouns are linked to more than one hypernym and can therefore cause structural damage on a lexical taxonomy. For instance, the Spanish noun tarántula ('tarantula') is a hyponym of araña ('spider'), but this is also a polysemous noun, as it means 'chandelier' as well. It is thus necessary to determine the next hypernym in the chain, that is animal ('animal') or artefacto ('artifact'). In this paper we explore methods to solve this problem using a similarity measure that uses verb-noun co-occurrence as a predictor variable. Best results (84 % success) are obtained with a simple method that only measures co-occurrence, irrespective of any syntactic information.

KW - Distributional similarity

KW - Hypernymy

KW - Polysemy

KW - Taxonomy

UR - http://www.scopus.com/inward/record.url?scp=85088387540&partnerID=8YFLogxK

U2 - 10.26342/2020-64-3

DO - 10.26342/2020-64-3

M3 - Article

AN - SCOPUS:85088387540

SN - 1135-5948

VL - 64

SP - 29

EP - 36

JO - Procesamiento de Lenguaje Natural

JF - Procesamiento de Lenguaje Natural

ER -

Tarántula -> araña -> Animal: Asignación de hiperónimos de segundo nivel basada en métodos de similitud distribucional

Abstract

Access to Document

Other files and links

Fingerprint

Cite this