Computational creativity is a subfield of artificial intelligence concerned with the development of programs that can produce creative output; in particular, several of these programs deal with linguistic creativity. Many computational creativity systems are modeled after, or inspired by, psychological and cognitive theories of creativity. One theory that accounts for the aesthetics of creative productions is summarized in the optimal innovation hypothesis, which claims that particular minimal modifications are more pleasurable than totally novel creations. Three systems based on this theory are presented; they creatively modify linguistic expressions with a pragmatic goal, such as attracting the attention of readers and evoking the contents of a news article or producing song parodies to help the listener remember a concept. The potential of this kind of systems is large: when paired with rich user data, they could produce creative messages based on what specific users are familiar with and achieve “personalized creativity.”

Cognition and computational linguistic creativity

Lorenzo Gatti;Oliviero Stock;Carlo Strapparava
2022-01-01

Abstract

Computational creativity is a subfield of artificial intelligence concerned with the development of programs that can produce creative output; in particular, several of these programs deal with linguistic creativity. Many computational creativity systems are modeled after, or inspired by, psychological and cognitive theories of creativity. One theory that accounts for the aesthetics of creative productions is summarized in the optimal innovation hypothesis, which claims that particular minimal modifications are more pleasurable than totally novel creations. Three systems based on this theory are presented; they creatively modify linguistic expressions with a pragmatic goal, such as attracting the attention of readers and evoking the contents of a news article or producing song parodies to help the listener remember a concept. The potential of this kind of systems is large: when paired with rich user data, they could produce creative messages based on what specific users are familiar with and achieve “personalized creativity.”
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11582/334567
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
social impact