Task Overview


Irony is a complex linguistic phenomenon widely studied in philosophy and linguistics. Irony can be defined as an incongruity between the literal meaning of an utterance and its intended meaning, that is meaning that is not strictly the conventional meaning of the individual words in the figurative expression. Irony overlaps with a variety of other figurative devices such as satire, parody, and sarcasm (Clark and Gerrig 1984; Gibbs 2000). In this task, irony is used as an umbrella term that includes sarcasm.

Irony detection has gained relevance recently, due to its importance in various NLP applications such as sentiment analysis, hate speech detection, author profiling, fake news detection, and crisis management (e.g., terrorist attacks, public disorder). For example, recent studies on irony show that the performances of sentiment analysis systems drastically decrease when applied to ironic texts (Benamara et al, 2017, Hernandez Faŕıas et al., 2016, 2018, Zhang et al, 2019). This is mainly due to the complexity of ironic contents that make use of figures of speech to convey non-literal meaning.

The task aims at detecting irony in Arabic tweets. Given a tweet, systems have to classify it as either ironic or not ironic. As far as we know, this is the first shared task on irony for the Arabic language.

Irony in the Arabic Language

Detecting irony in Arabic tweets poses a significant challenge, as Arabic tweets are often characterized by :

  • (a) non-diacritised texts,
  • (b) a large variations of unstandardized dialectal Arabic, and
  • (c) linguistic code switching between Modern Standard Arabic and several dialects and between Arabic and other languages like English and French.

The following two tweets illustrate this.

In tweet (1), the author ironically employs several positive opinion words ( ”character of the year”,   ”perfect”) towards Morsi, the former president of Egypt to express a clearly negative opinion. The tweet (2) is also negative towards Salim Cheboub, daughter’s husband of the former Tunsian president Ben Ali, although there are any negative opinion words. The author here describes an ironic situation due to a sudden change of topic (Ali Ben Salem’s Uncle vs. Salim Cheboub) and a false assertion (”Salim Cheboub says that he has no relationship with Ben Ali” is not true in reality).

This task will be a good opportunity to compare the performances of Arabic irony detection to those reported in recent shared tasks in other languages like English (Van Hee et al, 2018), French (Benamara et al, 2017), and Italian (Cignarella et al, 2018).

Target Audience

The target audience are researchers from public entities (Universities, Research Centers) or private companies who are interested in figurative language processing or enhancing the performances of other NLP applications like sentiment analysis, hate speech detection or author profiling.


Farah Benamara, Cyril Grouin, Jihen Karoui, Véronique Moriceau, Isabelle Robba. Introduction to the French irony Detetcion Shared Task (Analyse d’opinion et langage figuratif dans des tweets : présentation et résultats du Défi Fouille de Textes DEFT2017) DEFT@TALN2017. June 2017, Orléans, France.

Farah Benamara, Maite Taboada, Yvette Yannick Mathieu. Evaluative Language Beyond Bags of Words: Linguistic Insights and Computational Applications. Computational Linguistics 43(1): 201-264 (2017)

Alessandra Teresa Cignarella, Simona Frenda, Valerio Basile, Cristina Bosco, Viviana Patti, Paolo Rosso. Overview of the EVALITA 2018 Task on Irony Detection in Italian Tweets (IronITA). December 2018, Turin Italy.

Clark, H. H., Gerrig, R. J.. On the pretense theory of irony. Journal of Experimental Psychology: General 113 (1), 121–126. 1984

Gibbs, R. W.. Irony in talk among friends. Metaphor and symbol 15 (1-2), 5–27. 2000

Delia Irazú Hernandez, Paolo Rosso. Irony, Sarcasm, and Sentiment Analysis. Elsevier Science and Technology, Ch. 7 In: Sentiment Analysis in
Social Networks, pp. 113–128. 2018
Delia Irazú Hernández, Viviana Patti, Paolo Rosso. Irony Detection in Twitter: The Role of Affective Content. In: ACM Transactions on Internet Technology, vol. 16, issue 3, pp. 1-24, 2016
Cynthia Van Hee, Els Lefever, Véronique Hoste. SemEval-2018 Task 3: Irony Detection in English Tweets. proceedings of the SemEval shared tasks. June 2018, New Orleans, USA.
Shiwei Zhang, Xiuzhen Zhang, Jeffrey Chan, Paolo Rosso. Irony detection via sentiment-based transfer learning, Information Processing & Management, 56(5): 1633-1644, 2019. https://doi.org/10.1016/j.ipm.2019.04.006