On task effects in NLG corpus elicitation: A replication study using mixed effects modeling

Emiel van Miltenburg; Merel van de Kerkhof; Ruud Koolen; Martijn Goudbeek; Emiel Krahmer

On task effects in NLG corpus elicitation: A replication study using mixed effects modeling

Emiel van Miltenburg, Merel van de Kerkhof, Ruud Koolen, Martijn Goudbeek, Emiel Krahmer

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

83 Downloads (Pure)

Abstract

Task effects in NLG corpus elicitation recently started to receive more attention, but are usu- ally not modeled statistically. We present a controlled replication of the study by Van Mil- tenburg et al. (2018b), contrasting spoken with written descriptions. We collected additional written Dutch descriptions to supplement the spoken data from the DIDEC corpus, and an- alyzed the descriptions using mixed effects modeling to account for variation between par- ticipants and items. Our results show that the effects of modality largely disappear in a con- trolled setting.

Original language	English
Title of host publication	Proceedings of the 12th International Conference on Natural Language Generation (INLG 2019)
Editors	Kees van Deemter, Chenghua Lin, Hiroya Takamura
Publication status	Published - 2019
Event	12th International conference on Natural Language Generation (INLG 2019) - Tokyo, Japan Duration: 29 Oct 2019 → 1 Nov 2019 https://www.inlg2019.com

Conference

Conference	12th International conference on Natural Language Generation (INLG 2019)
Country/Territory	Japan
City	Tokyo
Period	29/10/19 → 1/11/19
Internet address	https://www.inlg2019.com

Keywords

Natural Language Generation
Image Description
Modality

Access to Document

Van Miltenburg et al. (2019) - INLGFinal published version, 159 KB

Cite this

@inproceedings{122b84548e3e442d9b16df840b36cc02,

title = "On task effects in NLG corpus elicitation: A replication study using mixed effects modeling",

abstract = "Task effects in NLG corpus elicitation recently started to receive more attention, but are usu- ally not modeled statistically. We present a controlled replication of the study by Van Mil- tenburg et al. (2018b), contrasting spoken with written descriptions. We collected additional written Dutch descriptions to supplement the spoken data from the DIDEC corpus, and an- alyzed the descriptions using mixed effects modeling to account for variation between par- ticipants and items. Our results show that the effects of modality largely disappear in a con- trolled setting.",

keywords = "Natural Language Generation, Image Description, Modality",

author = "{van Miltenburg}, Emiel and {van de Kerkhof}, Merel and Ruud Koolen and Martijn Goudbeek and Emiel Krahmer",

year = "2019",

language = "English",

editor = "{van Deemter}, Kees and Chenghua Lin and Hiroya Takamura",

booktitle = "Proceedings of the 12th International Conference on Natural Language Generation (INLG 2019)",

note = "12th International conference on Natural Language Generation (INLG 2019) ; Conference date: 29-10-2019 Through 01-11-2019",

url = "https://www.inlg2019.com",

}

van Miltenburg, E, van de Kerkhof, M, Koolen, R, Goudbeek, M & Krahmer, E 2019, On task effects in NLG corpus elicitation: A replication study using mixed effects modeling. in K van Deemter, C Lin & H Takamura (eds), Proceedings of the 12th International Conference on Natural Language Generation (INLG 2019). 12th International conference on Natural Language Generation (INLG 2019), Tokyo, Japan, 29/10/19.

On task effects in NLG corpus elicitation: A replication study using mixed effects modeling. / van Miltenburg, Emiel; van de Kerkhof, Merel; Koolen, Ruud et al.
Proceedings of the 12th International Conference on Natural Language Generation (INLG 2019). ed. / Kees van Deemter; Chenghua Lin; Hiroya Takamura. 2019.

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

TY - GEN

T1 - On task effects in NLG corpus elicitation

T2 - 12th International conference on Natural Language Generation (INLG 2019)

AU - van Miltenburg, Emiel

AU - van de Kerkhof, Merel

AU - Koolen, Ruud

AU - Goudbeek, Martijn

AU - Krahmer, Emiel

PY - 2019

Y1 - 2019

N2 - Task effects in NLG corpus elicitation recently started to receive more attention, but are usu- ally not modeled statistically. We present a controlled replication of the study by Van Mil- tenburg et al. (2018b), contrasting spoken with written descriptions. We collected additional written Dutch descriptions to supplement the spoken data from the DIDEC corpus, and an- alyzed the descriptions using mixed effects modeling to account for variation between par- ticipants and items. Our results show that the effects of modality largely disappear in a con- trolled setting.

AB - Task effects in NLG corpus elicitation recently started to receive more attention, but are usu- ally not modeled statistically. We present a controlled replication of the study by Van Mil- tenburg et al. (2018b), contrasting spoken with written descriptions. We collected additional written Dutch descriptions to supplement the spoken data from the DIDEC corpus, and an- alyzed the descriptions using mixed effects modeling to account for variation between par- ticipants and items. Our results show that the effects of modality largely disappear in a con- trolled setting.

KW - Natural Language Generation

KW - Image Description

KW - Modality

M3 - Conference contribution

BT - Proceedings of the 12th International Conference on Natural Language Generation (INLG 2019)

A2 - van Deemter, Kees

A2 - Lin, Chenghua

A2 - Takamura, Hiroya

Y2 - 29 October 2019 through 1 November 2019

ER -

On task effects in NLG corpus elicitation: A replication study using mixed effects modeling

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this