Lessons from Computational Modelling of Reference Production in Mandarin and English

Guanyi Chen; Kees van Deemter

Lessons from Computational Modelling of Reference Production in Mandarin and English

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

Abstract

Referring expression generation (REG) algorithms offer computational models of the production of referring expressions. In earlier work, a corpus of referring expressions (REs) in Mandarin was introduced. In the present paper, we annotate this corpus, evaluate classic REG algorithms on it, and compare the results with earlier results on the evaluation of REG for English referring expressions. Next, we offer an in-depth analysis of the corpus, focusing on issues that arise from the grammar of Mandarin. We discuss shortcomings of previous REG evaluations that came to light during our investigation and we highlight some surprising results. Perhaps most strikingly, we found a much higher proportion of under-specified expressions than previous studies had suggested, not just in Mandarin but in English as well.

Original language	English
Title of host publication	Proceedings of the 13th International Conference on Natural Language Generation
Editors	Brian Davis, Yvette Graham, John Kelleher, Yaji Sripada
Place of Publication	Dublin, Ireland
Publisher	Association for Computational Linguistics
Pages	263-272
Number of pages	10
Publication status	Published - 1 Dec 2020

Access to Document

2020.inlg-1.33v2Final published version, 411 KBLicence: Taverne

https://www.aclweb.org/anthology/2020.inlg-1.33

Cite this

@inproceedings{e6f6774009e245ce84ffa1dd479ecef6,

title = "Lessons from Computational Modelling of Reference Production in Mandarin and English",

abstract = "Referring expression generation (REG) algorithms offer computational models of the production of referring expressions. In earlier work, a corpus of referring expressions (REs) in Mandarin was introduced. In the present paper, we annotate this corpus, evaluate classic REG algorithms on it, and compare the results with earlier results on the evaluation of REG for English referring expressions. Next, we offer an in-depth analysis of the corpus, focusing on issues that arise from the grammar of Mandarin. We discuss shortcomings of previous REG evaluations that came to light during our investigation and we highlight some surprising results. Perhaps most strikingly, we found a much higher proportion of under-specified expressions than previous studies had suggested, not just in Mandarin but in English as well.",

author = "Guanyi Chen and {van Deemter}, Kees",

year = "2020",

month = dec,

day = "1",

language = "English",

pages = "263--272",

editor = "Davis, {Brian } and Graham, {Yvette } and Kelleher, {John } and Sripada, {Yaji }",

booktitle = "Proceedings of the 13th International Conference on Natural Language Generation",

publisher = "Association for Computational Linguistics",

address = "United States",

}

Chen, G & van Deemter, K 2020, Lessons from Computational Modelling of Reference Production in Mandarin and English. in B Davis, Y Graham, J Kelleher & Y Sripada (eds), Proceedings of the 13th International Conference on Natural Language Generation. Association for Computational Linguistics, Dublin, Ireland, pp. 263-272. <https://www.aclweb.org/anthology/2020.inlg-1.33>

Lessons from Computational Modelling of Reference Production in Mandarin and English. / Chen, Guanyi; van Deemter, Kees.
Proceedings of the 13th International Conference on Natural Language Generation. ed. / Brian Davis; Yvette Graham; John Kelleher; Yaji Sripada. Dublin, Ireland: Association for Computational Linguistics, 2020. p. 263-272.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Lessons from Computational Modelling of Reference Production in Mandarin and English

AU - Chen, Guanyi

AU - van Deemter, Kees

PY - 2020/12/1

Y1 - 2020/12/1

N2 - Referring expression generation (REG) algorithms offer computational models of the production of referring expressions. In earlier work, a corpus of referring expressions (REs) in Mandarin was introduced. In the present paper, we annotate this corpus, evaluate classic REG algorithms on it, and compare the results with earlier results on the evaluation of REG for English referring expressions. Next, we offer an in-depth analysis of the corpus, focusing on issues that arise from the grammar of Mandarin. We discuss shortcomings of previous REG evaluations that came to light during our investigation and we highlight some surprising results. Perhaps most strikingly, we found a much higher proportion of under-specified expressions than previous studies had suggested, not just in Mandarin but in English as well.

AB - Referring expression generation (REG) algorithms offer computational models of the production of referring expressions. In earlier work, a corpus of referring expressions (REs) in Mandarin was introduced. In the present paper, we annotate this corpus, evaluate classic REG algorithms on it, and compare the results with earlier results on the evaluation of REG for English referring expressions. Next, we offer an in-depth analysis of the corpus, focusing on issues that arise from the grammar of Mandarin. We discuss shortcomings of previous REG evaluations that came to light during our investigation and we highlight some surprising results. Perhaps most strikingly, we found a much higher proportion of under-specified expressions than previous studies had suggested, not just in Mandarin but in English as well.

M3 - Conference contribution

SP - 263

EP - 272

BT - Proceedings of the 13th International Conference on Natural Language Generation

A2 - Davis, Brian

A2 - Graham, Yvette

A2 - Kelleher, John

A2 - Sripada, Yaji

PB - Association for Computational Linguistics

CY - Dublin, Ireland

ER -

Lessons from Computational Modelling of Reference Production in Mandarin and English

Abstract

Access to Document

Fingerprint

Cite this