Exploiting Open IE for Deriving Multiple Premises Entailment
Corpus

D 2019

Exploiting Open IE for Deriving Multiple Premises Entailment Corpus

VÍTA, Martin a Jakub KLÍMEK

Základní údaje

Originální název

Exploiting Open IE for Deriving Multiple Premises Entailment Corpus

Autoři

VÍTA, Martin (203 Česká republika, garant, domácí) a Jakub KLÍMEK

Vydání

Varna, Proceedings of Recent Advances in Natural Language Processing, od s. 1257-1264, 8 s. 2019

Nakladatel

2019

Další údaje

Jazyk

angličtina

Typ výsledku

Stať ve sborníku

Obor

10201 Computer sciences, information science, bioinformatics

Stát vydavatele

Rumunsko

Utajení

není předmětem státního či obchodního tajemství

Forma vydání

elektronická verze "online"

Odkazy

Plný text sborníku

Kód RIV

RIV/00216224:14330/19:00110584

Organizační jednotka

Fakulta informatiky

ISBN

978-954-452-055-7

ISSN

DOI

http://dx.doi.org/10.26615/978-954-452-056-4_144

Klíčová slova anglicky

NLI; textual entailment; multiple premises entailment; open information extraction

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 11. 5. 2020 07:12, RNDr. Pavel Šmerk, Ph.D.

Anotace

V originále

Natural language inference (NLI) is a key part of natural language understanding. The NLI task is defined as a decision problem whether a given sentence -- hypothesis -- can be inferred from a given text. Typically, we deal with a text consisting of just a single premise/single sentence, which is called a single premise entailment (SPE) task. Recently, a derived task of NLI from multiple premises (MPE) was introduced together with the first annotated corpus and corresponding several strong baselines. Nevertheless, the further development in MPE field requires accessibility of huge amounts of annotated data. In this paper we introduce a novel method for rapid deriving of MPE corpora from an existing NLI (SPE) annotated data that does not require any additional annotation work. This proposed approach is based on using an open information extraction system. We demonstrate the application of the method on a well known SNLI corpus. Over the obtained corpus, we provide the first evaluations as well as we state a strong baseline.

Návaznosti

MUNI/A/1018/2018, interní kód MU

Název: Rozsáhlé výpočetní systémy: modely, aplikace a verifikace VIII.

Investor: Masarykova univerzita, Rozsáhlé výpočetní systémy: modely, aplikace a verifikace VIII., DO R. 2020_Kategorie A - Specifický výzkum - Studentské výzkumné projekty

Citovat

VÍTA, Martin a Jakub KLÍMEK. Exploiting Open IE for Deriving Multiple Premises Entailment Corpus. Online. In Galia Angelova, Ruslan Mitkov, Ivelina Nikolova, Irina Temnikova. Proceedings of Recent Advances in Natural Language Processing. Varna: 2019, 2019, s. 1257-1264. ISBN 978-954-452-055-7. Dostupné z: https://dx.doi.org/10.26615/978-954-452-056-4_144.

@inproceedings{1552616,
   author = {Víta, Martin and Klímek, Jakub},
   address = {Varna},
   booktitle = {Proceedings of Recent Advances in Natural Language Processing},
   doi = {http://dx.doi.org/10.26615/978-954-452-056-4_144},
   editor = {Galia Angelova, Ruslan Mitkov, Ivelina Nikolova, Irina Temnikova},
   keywords = {NLI; textual entailment; multiple premises entailment; open information extraction},
   howpublished = {elektronická verze "online"},
   language = {eng},
   location = {Varna},
   isbn = {978-954-452-055-7},
   pages = {1257-1264},
   publisher = {2019},
   title = {Exploiting Open IE for Deriving Multiple Premises Entailment Corpus},
   url = {http://lml.bas.bg/ranlp2019/proceedings-ranlp-2019.pdf},
   year = {2019}
}

TY  - JOUR
ID  - 1552616
AU  - Víta, Martin - Klímek, Jakub
PY  - 2019
TI  - Exploiting Open IE for Deriving Multiple Premises Entailment Corpus
PB  - 2019
CY  - Varna
SN  - 9789544520557
KW  - NLI
KW  - textual entailment
KW  - multiple premises entailment
KW  - open information extraction
UR  - http://lml.bas.bg/ranlp2019/proceedings-ranlp-2019.pdf
L2  - http://lml.bas.bg/ranlp2019/proceedings-ranlp-2019.pdf
N2  - Natural language inference (NLI) is a key part of natural language understanding. The NLI task is defined as a decision problem whether a given sentence -- hypothesis -- can be inferred from a given text. Typically, we deal with a text consisting of just a single premise/single sentence, which is called a single premise entailment (SPE) task. Recently, a derived task of NLI from multiple premises (MPE) was introduced together with the first annotated corpus and corresponding several strong baselines. Nevertheless, the further development in MPE field requires accessibility of huge amounts of annotated data. In this paper we introduce a novel method for rapid deriving of MPE corpora from an existing NLI (SPE) annotated data that does not require any additional annotation work. This proposed approach is based on using an open information extraction system. We demonstrate the application of the method on a well known SNLI corpus. Over the obtained corpus, we provide the first evaluations as well as we state a strong baseline.
ER  -

VÍTA, Martin a Jakub KLÍMEK. Exploiting Open IE for Deriving Multiple Premises Entailment Corpus. Online. In Galia Angelova, Ruslan Mitkov, Ivelina Nikolova, Irina Temnikova. \textit{Proceedings of Recent Advances in Natural Language Processing}. Varna: 2019, 2019, s.~1257-1264. ISBN~978-954-452-055-7. Dostupné z: https://dx.doi.org/10.26615/978-954-452-056-4\_{}144.

Podrobný výpis o publikaci