When Sarcasm Hurts: Irony-Aware Models for Abusive Language Detection

Simona Frenda; Viviana Patti; Paolo Rosso

doi:10.1007/978-3-031-42448-9_4

When Sarcasm Hurts: Irony-Aware Models for Abusive Language Detection

Simona Frenda^*
, Viviana Patti
, Paolo Rosso

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

10 Citations (Scopus)

Abstract

Linguistic literature on irony discusses sarcasm as a form of irony characterized by its biting nature and the intention to mock a victim. This particular trait makes sarcasm apt to convey hate speech and not only humour. Previous works on abusive language stressed the need to address ironic language to lead the system to recognize correctly hate speech, especially in spontaneous texts, like tweets [13]. In this context, our main hypothesis is that information about the presence of sarcasm could help to improve the detection of hateful messages, especially when they are camouflaged as sarcastic. To corroborate this hypothesis: i) we perform analysis on HaSpeeDe20_ext, an Italian corpus of tweets about the integration of cultural minorities in Italy, ii) we carry out computational experiments injecting the knowledge of sarcasm in a system of hate speech detection, and iii) we adopt strategies of validation in terms of performance and significance of the obtained results. Results confirm our hypothesis and overcome the state of the art.

Original language	English
Title of host publication	Experimental IR Meets Multilinguality, Multimodality, and Interaction
Subtitle of host publication	14th International Conference of the CLEF Association, CLEF 2023, Thessaloniki, Greece, September 18–21, 2023, Proceedings
Editors	Avi Arampatzis, Evangelos Kanoulas, Mohammad Aliannejadi, Theodora Tsikrika, Stefanos Vrochidis, Anastasia Giachanou, Dan Li, Michalis Vlachos, Guglielmo Faggioli, Nicola Ferro
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	34-47
Number of pages	14
ISBN (Electronic)	9783031424489
ISBN (Print)	9783031424472
DOIs	https://doi.org/10.1007/978-3-031-42448-9_4
Publication status	Published - 12 Sept 2023
Event	14th International Conference of the CLEF Association 2023 - Thessaloniki, Greece Duration: 18 Sept 2023 → 21 Sept 2023 https://clef2023.clef-initiative.eu/index.php?page=Pages/cfp.html

Publication series

Name	Lecture Notes in Computer Science
Volume	14163 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	14th International Conference of the CLEF Association 2023
Abbreviated title	CLEF 2023
Country/Territory	Greece
City	Thessaloniki
Period	18/09/23 → 21/09/23
Internet address	https://clef2023.clef-initiative.eu/index.php?page=Pages/cfp.html

Keywords

Abusive Language Detection
Irony Detection
Multi-Task Learning

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-031-42448-9_4

Cite this

Frenda, S., Patti, V., & Rosso, P. (2023). When Sarcasm Hurts: Irony-Aware Models for Abusive Language Detection. In A. Arampatzis, E. Kanoulas, M. Aliannejadi, T. Tsikrika, S. Vrochidis, A. Giachanou, D. Li, M. Vlachos, G. Faggioli, & N. Ferro (Eds.), Experimental IR Meets Multilinguality, Multimodality, and Interaction: 14th International Conference of the CLEF Association, CLEF 2023, Thessaloniki, Greece, September 18–21, 2023, Proceedings (pp. 34-47). (Lecture Notes in Computer Science; Vol. 14163 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-42448-9_4

Frenda, Simona ; Patti, Viviana ; Rosso, Paolo. / When Sarcasm Hurts: Irony-Aware Models for Abusive Language Detection. Experimental IR Meets Multilinguality, Multimodality, and Interaction: 14th International Conference of the CLEF Association, CLEF 2023, Thessaloniki, Greece, September 18–21, 2023, Proceedings. editor / Avi Arampatzis ; Evangelos Kanoulas ; Mohammad Aliannejadi ; Theodora Tsikrika ; Stefanos Vrochidis ; Anastasia Giachanou ; Dan Li ; Michalis Vlachos ; Guglielmo Faggioli ; Nicola Ferro. Springer Science and Business Media Deutschland GmbH, 2023. pp. 34-47 (Lecture Notes in Computer Science).

@inproceedings{d043e95579674af3bb8be601026e0b0b,

title = "When Sarcasm Hurts: Irony-Aware Models for Abusive Language Detection",

abstract = "Linguistic literature on irony discusses sarcasm as a form of irony characterized by its biting nature and the intention to mock a victim. This particular trait makes sarcasm apt to convey hate speech and not only humour. Previous works on abusive language stressed the need to address ironic language to lead the system to recognize correctly hate speech, especially in spontaneous texts, like tweets [13]. In this context, our main hypothesis is that information about the presence of sarcasm could help to improve the detection of hateful messages, especially when they are camouflaged as sarcastic. To corroborate this hypothesis: i) we perform analysis on HaSpeeDe20\_ext, an Italian corpus of tweets about the integration of cultural minorities in Italy, ii) we carry out computational experiments injecting the knowledge of sarcasm in a system of hate speech detection, and iii) we adopt strategies of validation in terms of performance and significance of the obtained results. Results confirm our hypothesis and overcome the state of the art.",

keywords = "Abusive Language Detection, Irony Detection, Multi-Task Learning",

author = "Simona Frenda and Viviana Patti and Paolo Rosso",

note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.; 14th International Conference of the CLEF Association 2023, CLEF 2023 ; Conference date: 18-09-2023 Through 21-09-2023",

year = "2023",

month = sep,

day = "12",

doi = "10.1007/978-3-031-42448-9\_4",

language = "English",

isbn = "9783031424472",

series = "Lecture Notes in Computer Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "34--47",

editor = "Avi Arampatzis and Evangelos Kanoulas and Mohammad Aliannejadi and Theodora Tsikrika and Stefanos Vrochidis and Anastasia Giachanou and Dan Li and Michalis Vlachos and Guglielmo Faggioli and Nicola Ferro",

booktitle = "Experimental IR Meets Multilinguality, Multimodality, and Interaction",

address = "Germany",

url = "https://clef2023.clef-initiative.eu/index.php?page=Pages/cfp.html",

}

Frenda, S, Patti, V & Rosso, P 2023, When Sarcasm Hurts: Irony-Aware Models for Abusive Language Detection. in A Arampatzis, E Kanoulas, M Aliannejadi, T Tsikrika, S Vrochidis, A Giachanou, D Li, M Vlachos, G Faggioli & N Ferro (eds), Experimental IR Meets Multilinguality, Multimodality, and Interaction: 14th International Conference of the CLEF Association, CLEF 2023, Thessaloniki, Greece, September 18–21, 2023, Proceedings. Lecture Notes in Computer Science, vol. 14163 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 34-47, 14th International Conference of the CLEF Association 2023, Thessaloniki, Greece, 18/09/23. https://doi.org/10.1007/978-3-031-42448-9_4

When Sarcasm Hurts: Irony-Aware Models for Abusive Language Detection. / Frenda, Simona; Patti, Viviana; Rosso, Paolo.
Experimental IR Meets Multilinguality, Multimodality, and Interaction: 14th International Conference of the CLEF Association, CLEF 2023, Thessaloniki, Greece, September 18–21, 2023, Proceedings. ed. / Avi Arampatzis; Evangelos Kanoulas; Mohammad Aliannejadi; Theodora Tsikrika; Stefanos Vrochidis; Anastasia Giachanou; Dan Li; Michalis Vlachos; Guglielmo Faggioli; Nicola Ferro. Springer Science and Business Media Deutschland GmbH, 2023. p. 34-47 (Lecture Notes in Computer Science; Vol. 14163 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - When Sarcasm Hurts: Irony-Aware Models for Abusive Language Detection

AU - Frenda, Simona

AU - Patti, Viviana

AU - Rosso, Paolo

PY - 2023/9/12

Y1 - 2023/9/12

N2 - Linguistic literature on irony discusses sarcasm as a form of irony characterized by its biting nature and the intention to mock a victim. This particular trait makes sarcasm apt to convey hate speech and not only humour. Previous works on abusive language stressed the need to address ironic language to lead the system to recognize correctly hate speech, especially in spontaneous texts, like tweets [13]. In this context, our main hypothesis is that information about the presence of sarcasm could help to improve the detection of hateful messages, especially when they are camouflaged as sarcastic. To corroborate this hypothesis: i) we perform analysis on HaSpeeDe20_ext, an Italian corpus of tweets about the integration of cultural minorities in Italy, ii) we carry out computational experiments injecting the knowledge of sarcasm in a system of hate speech detection, and iii) we adopt strategies of validation in terms of performance and significance of the obtained results. Results confirm our hypothesis and overcome the state of the art.

AB - Linguistic literature on irony discusses sarcasm as a form of irony characterized by its biting nature and the intention to mock a victim. This particular trait makes sarcasm apt to convey hate speech and not only humour. Previous works on abusive language stressed the need to address ironic language to lead the system to recognize correctly hate speech, especially in spontaneous texts, like tweets [13]. In this context, our main hypothesis is that information about the presence of sarcasm could help to improve the detection of hateful messages, especially when they are camouflaged as sarcastic. To corroborate this hypothesis: i) we perform analysis on HaSpeeDe20_ext, an Italian corpus of tweets about the integration of cultural minorities in Italy, ii) we carry out computational experiments injecting the knowledge of sarcasm in a system of hate speech detection, and iii) we adopt strategies of validation in terms of performance and significance of the obtained results. Results confirm our hypothesis and overcome the state of the art.

KW - Abusive Language Detection

KW - Irony Detection

KW - Multi-Task Learning

UR - https://www.scopus.com/pages/publications/85172361941

U2 - 10.1007/978-3-031-42448-9_4

DO - 10.1007/978-3-031-42448-9_4

M3 - Conference contribution

AN - SCOPUS:85172361941

SN - 9783031424472

T3 - Lecture Notes in Computer Science

SP - 34

EP - 47

BT - Experimental IR Meets Multilinguality, Multimodality, and Interaction

A2 - Arampatzis, Avi

A2 - Kanoulas, Evangelos

A2 - Aliannejadi, Mohammad

A2 - Tsikrika, Theodora

A2 - Vrochidis, Stefanos

A2 - Giachanou, Anastasia

A2 - Li, Dan

A2 - Vlachos, Michalis

A2 - Faggioli, Guglielmo

A2 - Ferro, Nicola

PB - Springer Science and Business Media Deutschland GmbH

T2 - 14th International Conference of the CLEF Association 2023

Y2 - 18 September 2023 through 21 September 2023

ER -

Frenda S, Patti V, Rosso P. When Sarcasm Hurts: Irony-Aware Models for Abusive Language Detection. In Arampatzis A, Kanoulas E, Aliannejadi M, Tsikrika T, Vrochidis S, Giachanou A, Li D, Vlachos M, Faggioli G, Ferro N, editors, Experimental IR Meets Multilinguality, Multimodality, and Interaction: 14th International Conference of the CLEF Association, CLEF 2023, Thessaloniki, Greece, September 18–21, 2023, Proceedings. Springer Science and Business Media Deutschland GmbH. 2023. p. 34-47. (Lecture Notes in Computer Science). Epub 2023 Sept 11. doi: 10.1007/978-3-031-42448-9_4

When Sarcasm Hurts: Irony-Aware Models for Abusive Language Detection

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this