• DocumentCode
    3703308
  • Title

    Inducing an ironic effect in automated tweets

  • Author

    Alessandro Valitutti;Tony Veale

  • Author_Institution
    School of Computer Science and Informatics, University College Dublin, Belfield, Dublin D4, Ireland
  • fYear
    2015
  • Firstpage
    153
  • Lastpage
    159
  • Abstract
    Irony gives us a way to react creatively to disappointment. By allowing us to speak of a failed expectation as though it succeeded, irony stresses the naturalness of our expectation and the absurdity of its failure. The result of this playful use of language is a subtle valence shift as listeners are alerted to a gap between what is said and what is meant. But as irony is not without risks, speakers are often careful to signal an ironic intent with tone, body language, or if on Twitter, with the hashtag #irony. Yet given the subtlety of irony, we question the effectiveness of explicit marking, and empirically show how a stronger valence shift can be induced in automatically-generated creative tweets with more nuanced signals of irony.
  • Keywords
    "Semantics","Twitter","Web services","Context","Investment","Computer science","Informatics"
  • Publisher
    ieee
  • Conference_Titel
    Affective Computing and Intelligent Interaction (ACII), 2015 International Conference on
  • Electronic_ISBN
    2156-8111
  • Type

    conf

  • DOI
    10.1109/ACII.2015.7344565
  • Filename
    7344565