Data-Driven News Generation for Automated Journalism

Leo Leppänen; Myriam Munezero; Mark Granroth-Wilding; Hannu Toivonen

doi:10.18653/v1/w17-3528

Data-Driven News Generation for Automated Journalism

Leo Leppänen, Myriam Munezero, Mark Granroth-Wilding, Hannu Toivonen

Forskningsoutput: Kapitel i bok/rapport/konferenshandling › Konferensbidrag › Vetenskaplig › Peer review

Sammanfattning

Despite increasing amounts of data and ever improving natural language generation techniques, work on automated journalism is still relatively scarce. In this paper, we explore the field and challenges associated with building a journalistic natural language generation system. We present a set of requirements that should guide system design, including transparency, accuracy, modifiability and transferability. Guided by the requirements, we present a data-driven architecture for automated journalism that is largely domain and language independent. We illustrate its practical application in the production of news articles upon a user request about the 2017 Finnish municipal elections in three languages, demonstrating the successfulness of the data-driven, modular approach of the design. We then draw some lessons for future automated journalism.

Originalspråk	engelska
Titel på gästpublikation	The 10th International Natural Language Generation conference, Proceedings of the Conference
Antal sidor	10
Utgivningsort	Stroudsburg
Förlag	The Association for Computational Linguistics
Utgivningsdatum	4 sep 2017
Sidor	188-197
ISBN (tryckt)	978-1-945626-52-4
DOI	https://doi.org/10.18653/v1/w17-3528
Status	Publicerad - 4 sep 2017
MoE-publikationstyp	A4 Artikel i en konferenspublikation
Evenemang	International Conference on Natural Language Generation - Tilburg, Holland Varaktighet: 5 nov 2018 → 8 nov 2018 Konferensnummer: 11 https://inlg2018.uvt.nl/

Vetenskapsgrenar

113 Data- och informationsvetenskap

Åtkomst av dokument

10.18653/v1/w17-3528

http://www.aclweb.org/anthology/W17-3528

1 Slutfört

IA: Immersive Automation: automated storytelling, audience engagement and user experience in a news ecosystem
Linden, C., Moring, T. A., Backholm, K., Hujanen, J., Toivonen, H. T., Granroth-Wilding, M., Leppänen, L., Munezero, M., Haapanen, L. & Sirén-Heikel, S.
Viestintäalan tutkimussäätiö, Sanoma Media Finland Oy, Alma Media Oyj, Conmio Oy, Tekes, Svenska kulturfonden, KSF Media Ab, Keski Pohjanmaan Kirjapaino Oyj
01/12/2016 → 31/05/2018
Projekt: Forskningsprojekt

1 Arrangemang av och deltagande i konferens/workshop/kurs/seminarium

International Conference on Natural Language Generation
Leo Leppänen (Talare: Presentation)
4 sep 2017 → 7 sep 2017
Aktivitet: Typer för deltagande i eller organisering av evenemang › Arrangemang av och deltagande i konferens/workshop/kurs/seminarium

Citera det här

@inproceedings{d763834694f4424f863eed967250b4aa,

title = "Data-Driven News Generation for Automated Journalism",

abstract = "Despite increasing amounts of data and ever improving natural language generation techniques, work on automated journalism is still relatively scarce. In this paper, we explore the field and challenges associated with building a journalistic natural language generation system. We present a set of requirements that should guide system design, including transparency, accuracy, modifiability and transferability. Guided by the requirements, we present a data-driven architecture for automated journalism that is largely domain and language independent. We illustrate its practical application in the production of news articles upon a user request about the 2017 Finnish municipal elections in three languages, demonstrating the successfulness of the data-driven, modular approach of the design. We then draw some lessons for future automated journalism.",

keywords = "113 Computer and information sciences, natural language generation, natural language processing, news automation, computational creativity",

author = "Leo Lepp{\"a}nen and Myriam Munezero and Mark Granroth-Wilding and Hannu Toivonen",

year = "2017",

month = sep,

day = "4",

doi = "10.18653/v1/w17-3528",

language = "English",

isbn = "978-1-945626-52-4",

pages = "188--197",

booktitle = "The 10th International Natural Language Generation conference, Proceedings of the Conference",

publisher = "The Association for Computational Linguistics",

address = "United States",

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

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

}

Leppänen, L, Munezero, M, Granroth-Wilding, M & Toivonen, H 2017, Data-Driven News Generation for Automated Journalism. i The 10th International Natural Language Generation conference, Proceedings of the Conference. The Association for Computational Linguistics, Stroudsburg, s. 188-197, International Conference on Natural Language Generation, Holland, 05/11/2018. https://doi.org/10.18653/v1/w17-3528

Data-Driven News Generation for Automated Journalism. / Leppänen, Leo; Munezero, Myriam; Granroth-Wilding, Mark ; Toivonen, Hannu.

The 10th International Natural Language Generation conference, Proceedings of the Conference. Stroudsburg : The Association for Computational Linguistics, 2017. s. 188-197.

Forskningsoutput: Kapitel i bok/rapport/konferenshandling › Konferensbidrag › Vetenskaplig › Peer review

TY - GEN

T1 - Data-Driven News Generation for Automated Journalism

AU - Leppänen, Leo

AU - Munezero, Myriam

AU - Granroth-Wilding, Mark

AU - Toivonen, Hannu

N1 - Conference code: 11

PY - 2017/9/4

Y1 - 2017/9/4

N2 - Despite increasing amounts of data and ever improving natural language generation techniques, work on automated journalism is still relatively scarce. In this paper, we explore the field and challenges associated with building a journalistic natural language generation system. We present a set of requirements that should guide system design, including transparency, accuracy, modifiability and transferability. Guided by the requirements, we present a data-driven architecture for automated journalism that is largely domain and language independent. We illustrate its practical application in the production of news articles upon a user request about the 2017 Finnish municipal elections in three languages, demonstrating the successfulness of the data-driven, modular approach of the design. We then draw some lessons for future automated journalism.

AB - Despite increasing amounts of data and ever improving natural language generation techniques, work on automated journalism is still relatively scarce. In this paper, we explore the field and challenges associated with building a journalistic natural language generation system. We present a set of requirements that should guide system design, including transparency, accuracy, modifiability and transferability. Guided by the requirements, we present a data-driven architecture for automated journalism that is largely domain and language independent. We illustrate its practical application in the production of news articles upon a user request about the 2017 Finnish municipal elections in three languages, demonstrating the successfulness of the data-driven, modular approach of the design. We then draw some lessons for future automated journalism.

KW - 113 Computer and information sciences

KW - natural language generation

KW - natural language processing

KW - news automation

KW - computational creativity

U2 - 10.18653/v1/w17-3528

DO - 10.18653/v1/w17-3528

M3 - Conference contribution

SN - 978-1-945626-52-4

SP - 188

EP - 197

BT - The 10th International Natural Language Generation conference, Proceedings of the Conference

PB - The Association for Computational Linguistics

CY - Stroudsburg

T2 - International Conference on Natural Language Generation

Y2 - 5 November 2018 through 8 November 2018

ER -

Data-Driven News Generation for Automated Journalism

Sammanfattning

Vetenskapsgrenar

Åtkomst av dokument

IA: Immersive Automation: automated storytelling, audience engagement and user experience in a news ecosystem

International Conference on Natural Language Generation

Citera det här