Automated learning of templates for data-to-text generation: comparing rule-based, statistical and neural methods

Chris van der Lee; Emiel Krahmer; Sander Wubben

doi:10.18653/v1/W18-6504

Automated learning of templates for data-to-text generation: comparing rule-based, statistical and neural methods

Chris van der Lee, Emiel Krahmer, Sander Wubben

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

Abstract

The current study investigated novel techniques and methods for trainable approaches to data-to-text generation. Neural Machine Translation was explored for the conversion from data to text as well as the addition of extra templatization steps of the data input and text output in the conversion process. Evaluation using BLEU did not find the Neural Machine Translation technique to perform any better compared to rule-based or Statistical Machine Translation, and the templatization method seemed to perform similarly or sometimes worse compared to direct data-to-text conversion. However, the human evaluation metrics indicated that Neural Machine Translation yielded the highest quality output and that the templatization method was able to increase text quality in multiple situations.

Original language	English
Title of host publication	Proceedings of the 11th International Conference on Natural Language Generation
Publisher	Association for Computational Linguistics (ACL)
Pages	35- 45
Number of pages	10
DOIs	https://doi.org/10.18653/v1/W18-6504
Publication status	Published - 2018
Event	11th International Conference on Natural Language Generation - Tilburg University, Tilburg, Netherlands Duration: 5 Nov 2018 → 8 Nov 2018 Conference number: 11 https://inlg2018.uvt.nl/

Conference

Conference	11th International Conference on Natural Language Generation
Abbreviated title	INLG 2018
Country/Territory	Netherlands
City	Tilburg
Period	5/11/18 → 8/11/18
Internet address	https://inlg2018.uvt.nl/

Keywords

data-to-text generation
Neural Machine Translation
Templatization method

Access to Document

10.18653/v1/W18-6504

http://aclweb.org/anthology/W18-6504

Cite this

@inproceedings{d4c0250674be409db9434569fafdf359,

title = "Automated learning of templates for data-to-text generation: comparing rule-based, statistical and neural methods",

abstract = "The current study investigated novel techniques and methods for trainable approaches to data-to-text generation. Neural Machine Translation was explored for the conversion from data to text as well as the addition of extra templatization steps of the data input and text output in the conversion process. Evaluation using BLEU did not find the Neural Machine Translation technique to perform any better compared to rule-based or Statistical Machine Translation, and the templatization method seemed to perform similarly or sometimes worse compared to direct data-to-text conversion. However, the human evaluation metrics indicated that Neural Machine Translation yielded the highest quality output and that the templatization method was able to increase text quality in multiple situations.",

keywords = "data-to-text generation, Neural Machine Translation, Templatization method",

author = "\{van der Lee\}, Chris and Emiel Krahmer and Sander Wubben",

year = "2018",

doi = "10.18653/v1/W18-6504",

language = "English",

pages = "35-- 45",

booktitle = "Proceedings of the 11th International Conference on Natural Language Generation",

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

note = "11th International Conference on Natural Language Generation, INLG 2018 ; Conference date: 05-11-2018 Through 08-11-2018",

url = "https://inlg2018.uvt.nl/",

}

van der Lee, C , Krahmer, E & Wubben, S 2018, Automated learning of templates for data-to-text generation: comparing rule-based, statistical and neural methods. in Proceedings of the 11th International Conference on Natural Language Generation. Association for Computational Linguistics (ACL), pp. 35- 45, 11th International Conference on Natural Language Generation, Tilburg, Netherlands, 5/11/18. https://doi.org/10.18653/v1/W18-6504

Automated learning of templates for data-to-text generation: comparing rule-based, statistical and neural methods. / van der Lee, Chris ; Krahmer, Emiel; Wubben, Sander.
Proceedings of the 11th International Conference on Natural Language Generation. Association for Computational Linguistics (ACL), 2018. p. 35- 45.

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

TY - GEN

T1 - Automated learning of templates for data-to-text generation

T2 - 11th International Conference on Natural Language Generation

AU - van der Lee, Chris

AU - Krahmer, Emiel

AU - Wubben, Sander

N1 - Conference code: 11

PY - 2018

Y1 - 2018

N2 - The current study investigated novel techniques and methods for trainable approaches to data-to-text generation. Neural Machine Translation was explored for the conversion from data to text as well as the addition of extra templatization steps of the data input and text output in the conversion process. Evaluation using BLEU did not find the Neural Machine Translation technique to perform any better compared to rule-based or Statistical Machine Translation, and the templatization method seemed to perform similarly or sometimes worse compared to direct data-to-text conversion. However, the human evaluation metrics indicated that Neural Machine Translation yielded the highest quality output and that the templatization method was able to increase text quality in multiple situations.

AB - The current study investigated novel techniques and methods for trainable approaches to data-to-text generation. Neural Machine Translation was explored for the conversion from data to text as well as the addition of extra templatization steps of the data input and text output in the conversion process. Evaluation using BLEU did not find the Neural Machine Translation technique to perform any better compared to rule-based or Statistical Machine Translation, and the templatization method seemed to perform similarly or sometimes worse compared to direct data-to-text conversion. However, the human evaluation metrics indicated that Neural Machine Translation yielded the highest quality output and that the templatization method was able to increase text quality in multiple situations.

KW - data-to-text generation

KW - Neural Machine Translation

KW - Templatization method

U2 - 10.18653/v1/W18-6504

DO - 10.18653/v1/W18-6504

M3 - Conference contribution

SP - 35

EP - 45

BT - Proceedings of the 11th International Conference on Natural Language Generation

PB - Association for Computational Linguistics (ACL)

Y2 - 5 November 2018 through 8 November 2018

ER -

Automated learning of templates for data-to-text generation: comparing rule-based, statistical and neural methods

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this