A computational exploration of exaggeration

Enrica Troiano; Carlo Strapparava; Gözde Özbal; Serra Sinem Tekiroglu

A computational exploration of exaggeration

Enrica Troiano, Carlo Strapparava, Gözde Özbal, Serra Sinem Tekiroglu

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

Abstract

© 2018 Association for Computational LinguisticsSeveral NLP studies address the problem of figurative language, but among non-literal phenomena, they have neglected exaggeration. This paper presents a first computational approach to this figure of speech. We explore the possibility to automatically detect exaggerated sentences. First, we introduce HYPO, a corpus containing overstatements (or hyperboles) collected on the web and validated via crowdsourcing. Then, we evaluate a number of models trained on HYPO, and bring evidence that the task of hyperbole identification can be successfully performed based on a small set of semantic features.

Original language	English
Title of host publication	Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018
Editors	E. Riloff, D. Chiang, J. Hockenmaier, J. Tsujii
Publisher	Association for Computational Linguistics
Pages	3296-3304
ISBN (Electronic)	9781948087841
Publication status	Published - 2018
Externally published	Yes
Event	2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018 - Brussels, Belgium Duration: 31 Oct 2018 → 4 Nov 2018

Conference

Conference	2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018
Country/Territory	Belgium
City	Brussels
Period	31/10/18 → 4/11/18

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Persistent URL (handle)

Persistent link

Cite this

@inproceedings{bc10e37b9d9940ec89d2be5ae46889fa,

title = "A computational exploration of exaggeration",

abstract = "{\textcopyright} 2018 Association for Computational LinguisticsSeveral NLP studies address the problem of figurative language, but among non-literal phenomena, they have neglected exaggeration. This paper presents a first computational approach to this figure of speech. We explore the possibility to automatically detect exaggerated sentences. First, we introduce HYPO, a corpus containing overstatements (or hyperboles) collected on the web and validated via crowdsourcing. Then, we evaluate a number of models trained on HYPO, and bring evidence that the task of hyperbole identification can be successfully performed based on a small set of semantic features.",

author = "Enrica Troiano and Carlo Strapparava and G{\"o}zde {\"O}zbal and Tekiroglu, {Serra Sinem}",

year = "2018",

language = "English",

pages = "3296--3304",

editor = "E. Riloff and D. Chiang and J. Hockenmaier and J. Tsujii",

booktitle = "Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018",

publisher = "Association for Computational Linguistics",

note = "2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018 ; Conference date: 31-10-2018 Through 04-11-2018",

}

Troiano, E, Strapparava, C, Özbal, G & Tekiroglu, SS 2018, A computational exploration of exaggeration. in E Riloff, D Chiang, J Hockenmaier & J Tsujii (eds), Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018. Association for Computational Linguistics, pp. 3296-3304, 2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018, Brussels, Belgium, 31/10/18.

A computational exploration of exaggeration. / Troiano, Enrica; Strapparava, Carlo; Özbal, Gözde et al.
Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018. ed. / E. Riloff; D. Chiang; J. Hockenmaier; J. Tsujii. Association for Computational Linguistics, 2018. p. 3296-3304.

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

TY - GEN

T1 - A computational exploration of exaggeration

AU - Troiano, Enrica

AU - Strapparava, Carlo

AU - Özbal, Gözde

AU - Tekiroglu, Serra Sinem

PY - 2018

Y1 - 2018

N2 - © 2018 Association for Computational LinguisticsSeveral NLP studies address the problem of figurative language, but among non-literal phenomena, they have neglected exaggeration. This paper presents a first computational approach to this figure of speech. We explore the possibility to automatically detect exaggerated sentences. First, we introduce HYPO, a corpus containing overstatements (or hyperboles) collected on the web and validated via crowdsourcing. Then, we evaluate a number of models trained on HYPO, and bring evidence that the task of hyperbole identification can be successfully performed based on a small set of semantic features.

AB - © 2018 Association for Computational LinguisticsSeveral NLP studies address the problem of figurative language, but among non-literal phenomena, they have neglected exaggeration. This paper presents a first computational approach to this figure of speech. We explore the possibility to automatically detect exaggerated sentences. First, we introduce HYPO, a corpus containing overstatements (or hyperboles) collected on the web and validated via crowdsourcing. Then, we evaluate a number of models trained on HYPO, and bring evidence that the task of hyperbole identification can be successfully performed based on a small set of semantic features.

UR - http://www.scopus.com/inward/record.url?scp=85079496469&partnerID=8YFLogxK

M3 - Conference contribution

SP - 3296

EP - 3304

BT - Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018

A2 - Riloff, E.

A2 - Chiang, D.

A2 - Hockenmaier, J.

A2 - Tsujii, J.

PB - Association for Computational Linguistics

T2 - 2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018

Y2 - 31 October 2018 through 4 November 2018

ER -

A computational exploration of exaggeration

Abstract

Conference

UN SDGs

Persistent URL (handle)

Other files and links

Fingerprint

Cite this