AMR Alignment for Morphologically-rich and Pro-drop Languages

Elif Oral; Gülsen Eryigit

AMR Alignment for Morphologically-rich and Pro-drop Languages

Elif Oral, Gülsen Eryigit

Yapay Zeka ve Veri Mühendisliği Bölümü

Istanbul Technical University

Araştırma sonucu: ???type-name??? › Konferans katkısı › bilirkişi

2 Atıf (Scopus)

Özet

Alignment between concepts in an abstract meaning representation (AMR) graph and the words within a sentence is one of the important stages of AMR parsing. Although there exist high performing AMR aligners for English, unfortunately, these are not well suited for many languages where many concepts appear from morpho-semantic elements. For the first time in the literature, this paper presents an AMR aligner tailored for morphologically-rich and pro-drop languages by experimenting on the Turkish language being a prominent example of this language group. Our aligner focuses on the meaning considering the rich Turkish morphology and aligns AMR concepts that emerge from morphemes using a tree traversal approach without additional resources or rules. We evaluate our aligner over a manually annotated gold data set. Our aligner outperforms the Turkish adaptations of the previously proposed aligners for English and Portuguese by an F1 score of 0.87 and provides a relative error reduction of up to 76%.

Orijinal dil	İngilizce
Ana bilgisayar yayını başlığı	ACL 2022 - 60th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Student Research Workshop
Editörler	Samuel Louvan, Andrea Madotto, Brielen Madureira
Yayınlayan	Association for Computational Linguistics (ACL)
Sayfalar	143-152
Sayfa sayısı	10
ISBN (Elektronik)	9781955917230
Yayın durumu	Yayınlandı - 2022
Etkinlik	60th Annual Meeting of the Association for Computational Linguistics, ACL 2022 - Dublin, Ireland Süre: 22 May 2022 → 27 May 2022

Yayın serisi

Adı	Proceedings of the Annual Meeting of the Association for Computational Linguistics
ISSN (Basılı)	0736-587X

???event.eventtypes.event.conference???

???event.eventtypes.event.conference???	60th Annual Meeting of the Association for Computational Linguistics, ACL 2022
Ülke/Bölge	Ireland
Şehir	Dublin
Periyot	22/05/22 → 27/05/22

Bibliyografik not

Publisher Copyright:
© 2022 Association for Computational Linguistics.

Diğer Dosyalar ve Linkler

Link to publication in Scopus

Alıntı Yap

Oral, E., & Eryigit, G. (2022). AMR Alignment for Morphologically-rich and Pro-drop Languages. In S. Louvan, A. Madotto, & B. Madureira (Eds.), ACL 2022 - 60th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Student Research Workshop (pp. 143-152). (Proceedings of the Annual Meeting of the Association for Computational Linguistics). Association for Computational Linguistics (ACL).

Oral, Elif ; Eryigit, Gülsen. / AMR Alignment for Morphologically-rich and Pro-drop Languages. ACL 2022 - 60th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Student Research Workshop. editör / Samuel Louvan ; Andrea Madotto ; Brielen Madureira. Association for Computational Linguistics (ACL), 2022. pp. 143-152 (Proceedings of the Annual Meeting of the Association for Computational Linguistics).

@inproceedings{729a0db784ab4c059bad723174f4594f,

title = "AMR Alignment for Morphologically-rich and Pro-drop Languages",

abstract = "Alignment between concepts in an abstract meaning representation (AMR) graph and the words within a sentence is one of the important stages of AMR parsing. Although there exist high performing AMR aligners for English, unfortunately, these are not well suited for many languages where many concepts appear from morpho-semantic elements. For the first time in the literature, this paper presents an AMR aligner tailored for morphologically-rich and pro-drop languages by experimenting on the Turkish language being a prominent example of this language group. Our aligner focuses on the meaning considering the rich Turkish morphology and aligns AMR concepts that emerge from morphemes using a tree traversal approach without additional resources or rules. We evaluate our aligner over a manually annotated gold data set. Our aligner outperforms the Turkish adaptations of the previously proposed aligners for English and Portuguese by an F1 score of 0.87 and provides a relative error reduction of up to 76%.",

author = "Elif Oral and G{\"u}lsen Eryigit",

note = "Publisher Copyright: {\textcopyright} 2022 Association for Computational Linguistics.; 60th Annual Meeting of the Association for Computational Linguistics, ACL 2022 ; Conference date: 22-05-2022 Through 27-05-2022",

year = "2022",

language = "English",

series = "Proceedings of the Annual Meeting of the Association for Computational Linguistics",

publisher = "Association for Computational Linguistics (ACL)",

pages = "143--152",

editor = "Samuel Louvan and Andrea Madotto and Brielen Madureira",

booktitle = "ACL 2022 - 60th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Student Research Workshop",

address = "United States",

}

Oral, E & Eryigit, G 2022, AMR Alignment for Morphologically-rich and Pro-drop Languages. in S Louvan, A Madotto & B Madureira (eds), ACL 2022 - 60th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Student Research Workshop. Proceedings of the Annual Meeting of the Association for Computational Linguistics, Association for Computational Linguistics (ACL), pp. 143-152, 60th Annual Meeting of the Association for Computational Linguistics, ACL 2022, Dublin, Ireland, 22/05/22.

AMR Alignment for Morphologically-rich and Pro-drop Languages. / Oral, Elif; Eryigit, Gülsen.
ACL 2022 - 60th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Student Research Workshop. ed. / Samuel Louvan; Andrea Madotto; Brielen Madureira. Association for Computational Linguistics (ACL), 2022. p. 143-152 (Proceedings of the Annual Meeting of the Association for Computational Linguistics).

Araştırma sonucu: ???type-name??? › Konferans katkısı › bilirkişi

TY - GEN

T1 - AMR Alignment for Morphologically-rich and Pro-drop Languages

AU - Oral, Elif

AU - Eryigit, Gülsen

PY - 2022

Y1 - 2022

N2 - Alignment between concepts in an abstract meaning representation (AMR) graph and the words within a sentence is one of the important stages of AMR parsing. Although there exist high performing AMR aligners for English, unfortunately, these are not well suited for many languages where many concepts appear from morpho-semantic elements. For the first time in the literature, this paper presents an AMR aligner tailored for morphologically-rich and pro-drop languages by experimenting on the Turkish language being a prominent example of this language group. Our aligner focuses on the meaning considering the rich Turkish morphology and aligns AMR concepts that emerge from morphemes using a tree traversal approach without additional resources or rules. We evaluate our aligner over a manually annotated gold data set. Our aligner outperforms the Turkish adaptations of the previously proposed aligners for English and Portuguese by an F1 score of 0.87 and provides a relative error reduction of up to 76%.

AB - Alignment between concepts in an abstract meaning representation (AMR) graph and the words within a sentence is one of the important stages of AMR parsing. Although there exist high performing AMR aligners for English, unfortunately, these are not well suited for many languages where many concepts appear from morpho-semantic elements. For the first time in the literature, this paper presents an AMR aligner tailored for morphologically-rich and pro-drop languages by experimenting on the Turkish language being a prominent example of this language group. Our aligner focuses on the meaning considering the rich Turkish morphology and aligns AMR concepts that emerge from morphemes using a tree traversal approach without additional resources or rules. We evaluate our aligner over a manually annotated gold data set. Our aligner outperforms the Turkish adaptations of the previously proposed aligners for English and Portuguese by an F1 score of 0.87 and provides a relative error reduction of up to 76%.

UR - http://www.scopus.com/inward/record.url?scp=85149123758&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85149123758

T3 - Proceedings of the Annual Meeting of the Association for Computational Linguistics

SP - 143

EP - 152

BT - ACL 2022 - 60th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Student Research Workshop

A2 - Louvan, Samuel

A2 - Madotto, Andrea

A2 - Madureira, Brielen

PB - Association for Computational Linguistics (ACL)

T2 - 60th Annual Meeting of the Association for Computational Linguistics, ACL 2022

Y2 - 22 May 2022 through 27 May 2022

ER -

Oral E, Eryigit G. AMR Alignment for Morphologically-rich and Pro-drop Languages. In Louvan S, Madotto A, Madureira B, editörler, ACL 2022 - 60th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Student Research Workshop. Association for Computational Linguistics (ACL). 2022. p. 143-152. (Proceedings of the Annual Meeting of the Association for Computational Linguistics).

AMR Alignment for Morphologically-rich and Pro-drop Languages

Özet

Yayın serisi

???event.eventtypes.event.conference???

Bibliyografik not

Diğer Dosyalar ve Linkler

Parmak izi

Alıntı Yap