[1T3-OS-32a] OS-32

[1T3-OS-32a-04] Cross-Genre Hybrid Automatic Story Generation Based on Structure Analysis

〇Hajime Murai¹, Mizuki Aoyama¹, Shoki Ohta¹, Takaki Fukumoto¹, Arisa Ohba¹, Yuni Saito¹, Eiichi Sato¹ (1. Future University Hakodate)

Keywords:Automatic Story Generation, Structure Analysis, Cross Genre

In order to realize cross-genre automatic story generation, a hybrid automatic story generation model was adopted. The hybrid model combines automatic story structure generation based on structure analysis for existing works, and text generation by a large language model. At first about 1500 highly rated Japanese entertainment stories were selected and analyzed from five genres, "Adventure", "Battle", "Horror", "Love", and "Detective". 17 story factors which correspond to frequently appeared story plots were statistically extracted. After that, a story structure system was developed utilizing 17 story factors. The resultant structures were converted to a prompt, and final plots were generated by a large language model. This system is able to generate stories that are understandable and reflecting extracted 17 story factors.