Authors
Jumpei Ono1, *, Takashi Ogata2
1
Department of IT Solution, Vocational School of Digital Arts Sendai, Sendai,
Miyagi 980-0003, Japan
2
Department of IT Solution, Faculty of Software and Information Science,
Iwate Prefectural University, Takizawa, Iwate 020-0693, Japan
*
Corresponding author. Email: [email protected]
Corresponding Author
Jumpei Ono
Received 3 June 2018, Accepted 30 September 2018, Available Online 1 December
2018.
DOI
https://doi.org/10.2991/jrnal.2018.5.3.15
Keywords
Case frame; integrated narrative generation system; knowledge acquisition;
random story generation; script; verb concept
Abstract
The integrated narrative generation system (INGS), that the authors have
developed, generates a story to translate the story into the surface representation.
In the story generation process, the integrated narrative generation system
uses narrative knowledge that was automatically acquired from existing
narrative works. This paper presents a method to acquire short scripts,
which are a kind of narrative knowledge, from existing works in Aozora
Bunko for the story generation. This paper presents a mechanism to generate
random story-like event sequences by using 23,751,142 bi-gram scripts acquired
based on the method proposed below. The authors aim to use the scripts
generated by the method as a first set to be revised through the next learning
process.
Copyright
© 2018 The Authors. Published by ALife Robotics Corp. Ltd..
Open Access
This is an open access article under the CC BY-NC license (http://creativecommons.org/licences/by-nc/4.0/).