Humor logic mechanism classification

For a humorous tweet, this task is to predict the mechanism by which the tweet conveys humor from a predefined set of categories:

  • Absurd: Humor comes from a logical inconsistency in the reasoning.
  • Analogy: It is a comparison between dissimilar elements.
  • Embarrassment: In the punchline one of the participants shames or embarrasses another one.
  • Exaggeration: There is a situation or comparison that is exaggerated.
  • Insults: There are insults to the characters in the joke or to real life people.
  • Irony: They say something but mean the opposite, or they describe a contradictory situation.
  • Misunderstanding: Humor comes from a participant understanding a question or a situation wrong.
  • Parody: The text is similar to another known text or work (for example a song, a saying, or a movie dialog) but it is modified to make it humorous.
  • Reference: It describes a real life situation, generally mundane, that the reader might relate to or not, but when the reader does identify with the situation it results in a humorous effect .
  • Stereotype: Humor comes from using a social group, ethnicity or profession to remark on a stereotypical characteristic.
  • Unmasking: Humor comes from a character acting in a certain way and later showing that their intentions or characteristics were different than initially thought.
  • Wordplay: Uses word ambiguity, made up words or combinations of words to give a humorous sense.
Publication
Luis Chiruzzo, Santiago Castro, Santiago Góngora, Aiala Rosa, J. A. Meaney, Rada Mihalcean(2021) Overview of HAHA at IberLEF 2021: Detecting, Rating and Analyzing Humor in Spanish. Procesamiento del Lenguaje Natural, Revista nº 67, septiembre de 2021, pp. 257-268.
Language
Spanish
NLP topic
Abstract task
Dataset
Year
2021
Ranking metric
Macro F1

Task results

System MacroF1 Sort ascending
BERT4EVER 0.3396
Jocoso 0.2916
icc 0.2522
skblaz 0.2355
kuiyongyi 0.2187

If you have published a result better than those on the list, send a message to odesia-comunicacion@lsi.uned.es indicating the result and the DOI of the article, along with a copy of it if it is not published openly.