Host: The Japanese Society for Artificial Intelligence
Name : The 32nd Annual Conference of the Japanese Society for Artificial Intelligence, 2018
Number : 32
Location : [in Japanese]
Date : June 05, 2018 - June 08, 2018
The aim of this research is to develop a comic search system based on the characteristics of characters appearing in comics. For that purpose, this paper examines classification of characters using the egogram, which is used to classify personality. In the proposed method,texts that express the comic character's characteristics are acquired from Web resources and semantic vectors are allocated based on the texts using the egogram. The resulting egogram pattern is used to estimate the typical properties. Our experiment revealed that the performance of classification was 55.0 %.
