Figure of Speech Detection and Generation as a Service in IDN Authoring Support

Open Access
Authors
Publication date 2023
Host editors
  • L. Holloway-Attaway
  • J.T. Murray
Book title Interactive Storytelling
Book subtitle 16th International Conference on Interactive Digital Storytelling, ICIDS 2023, Kobe, Japan, November 11–15, 2023 : proceedings
ISBN
  • 9783031476570
ISBN (electronic)
  • 9783031476587
Series Lecture Notes in Computer Science
Event 16th International Conference on Interactive Digital Storytelling, ICIDS 2023
Volume | Issue number II
Pages (from-to) 99-112
Number of pages 14
Publisher Cham: Springer
Organisations
  • Faculty of Science (FNWI) - Informatics Institute (IVI)
Abstract

IDN authoring is a complex and creative endeavour. This paper provides the basis for a collaborative text-based authoring tool to support the work with figures of speech, such as metaphors, similes and metonymies. Several models have been tested and BERT, as best performing, has been used for creating functions to predict a figure of speech and, using an embedding of the custom database, return figures of speeches that are closest to the input, and generating new figures of speech. The implemented functions were validated using writers active in different fields. All the code, the databases and the models can be found in this GitHub: link to GitHub.

Document type Conference contribution
Language English
Published at https://doi.org/10.1007/978-3-031-47658-7_8
Other links https://github.com/Annony9/FigureOfSpeech/tree/main https://www.scopus.com/pages/publications/85177456437
Downloads
978-3-031-47658-7_8 (Final published version)
Permalink to this page
Back