A Comparison of Feature-Based and Neural Scansion of Poetry

Manex Agirrezabal; Iñaki Alegria; Mans Hulden

A Comparison of Feature-Based and Neural Scansion of Poetry

Manex Agirrezabal, Iñaki Alegria, Mans Hulden

Institut for Nordiske Studier og Sprogvidenskab

3 Citationer (Scopus)

Abstract

Automatic analysis of poetic rhythm is a challenging task that involves linguistics, literature, and computer science. When the language to be analyzed is known, rule-based systems or data-driven methods can be used. In this paper, we analyze poetic rhythm in English and Spanish. We show that the representations of data learned from character-based neural models are more informative than the ones from hand-crafted features, and that a Bi-LSTM+CRF-model produces state-of-the art accuracy on scansion of poetry in two languages. Results also show that the information about whole word structure, and not just independent syllables, is highly informative for performing scansion.

Originalsprog	Engelsk
Tidsskrift	Proceedings of the International Conference Recent Advances in Natural Language Processing, RANLP 2017
Sider (fra-til)	18-23
Antal sider	6
Status	Udgivet - 2017

Adgang til dokumentet

https://acl-bg.org/proceedings/2017/RANLP%202017/pdf/RANLP003.pdfLicens: CC BY-NC-ND

Citationsformater

@article{0e13869893bd4e21b7d06b2e21242d52,

title = "A Comparison of Feature-Based and Neural Scansion of Poetry",

abstract = "Automatic analysis of poetic rhythm is a challenging task that involves linguistics, literature, and computer science. When the language to be analyzed is known, rule-based systems or data-driven methods can be used. In this paper, we analyze poetic rhythm in English and Spanish. We show that the representations of data learned from character-based neural models are more informative than the ones from hand-crafted features, and that a Bi-LSTM+CRF-model produces state-of-the art accuracy on scansion of poetry in two languages. Results also show that the information about whole word structure, and not just independent syllables, is highly informative for performing scansion. ",

author = "Manex Agirrezabal and I{\~n}aki Alegria and Mans Hulden",

year = "2017",

language = "English",

pages = "18--23",

journal = "Proceedings of the International Conference Recent Advances in Natural Language Processing, RANLP 2017",

}

TY - JOUR

T1 - A Comparison of Feature-Based and Neural Scansion of Poetry

AU - Agirrezabal, Manex

AU - Alegria, Iñaki

AU - Hulden, Mans

PY - 2017

Y1 - 2017

N2 - Automatic analysis of poetic rhythm is a challenging task that involves linguistics, literature, and computer science. When the language to be analyzed is known, rule-based systems or data-driven methods can be used. In this paper, we analyze poetic rhythm in English and Spanish. We show that the representations of data learned from character-based neural models are more informative than the ones from hand-crafted features, and that a Bi-LSTM+CRF-model produces state-of-the art accuracy on scansion of poetry in two languages. Results also show that the information about whole word structure, and not just independent syllables, is highly informative for performing scansion.

AB - Automatic analysis of poetic rhythm is a challenging task that involves linguistics, literature, and computer science. When the language to be analyzed is known, rule-based systems or data-driven methods can be used. In this paper, we analyze poetic rhythm in English and Spanish. We show that the representations of data learned from character-based neural models are more informative than the ones from hand-crafted features, and that a Bi-LSTM+CRF-model produces state-of-the art accuracy on scansion of poetry in two languages. Results also show that the information about whole word structure, and not just independent syllables, is highly informative for performing scansion.

M3 - Journal article

SP - 18

EP - 23

JO - Proceedings of the International Conference Recent Advances in Natural Language Processing, RANLP 2017

JF - Proceedings of the International Conference Recent Advances in Natural Language Processing, RANLP 2017

ER -

A Comparison of Feature-Based and Neural Scansion of Poetry

Abstract

Adgang til dokumentet

Fingeraftryk

Citationsformater