クロカワ　ヒロアキ   KUROKAWA HIROAKI   黒川　弘章    所属   工学部 電気電子工学科    職種   教授
言語種別	日本語
発行・発表の年月	2018/03
形態種別	国際会議論文
査読	査読あり
標題	Characteristic extraction method of human’s strategy in the Rock-Paper-Scissors game
執筆形態	共著
掲載誌名	Proc. of NCSP18
巻・号・頁	592-595頁
著者・共著者	Takahiro Komai, Song-Ju Kim, Hiroaki Kurokawa
概要	In this study, we focus on the Rock-Paper-Scissors(RPS) games. the RPS game is one of the games that competes for its victory or defeat by using three kinds of hand signs in the three‐cornered deadlock relationship. We investigate a conventional method to infer the RPS game strategy based on player’s personality from player’s behavior. The conventional method based on the genetic programming is verified by simulations. From the simulation, the conventional method can provide effective results for various RPS strategies, but it is not effective in some cases. Considering the results of the conventional method, we propose the method to decide the hand sign(Rock, Paper, and Scissors) with the highest winning rate by counting the hand signs of the opponent and verify its effectiveness.