Mining Preferences on Identifying Werewolf Players from Werewolf Game Logs
The deception party game “Are You a Werewolf?” requires players to guess other’s roles through discussions that are based on one’s own role and other players’ crucial utterances. This paper proposes a method to mine the empirical preference data used to identify werewolves from game logs. This involves obtaining an empirical preference related to the practice of divination. In this method, if one of the three players revealing oneself as a seer divines a player as a human, while the other two players divine the player as a werewolf, then it can be judged that the divined player’s role is a werewolf.
KeywordsThe werewolf game “Are You a Werewolf?” Game logs Mining
- 1.Chittaranjan, G., Hung, H.: Are you a werewolf? detecting deceptive roles and outcomes in a conversational role-playing game. In: Proceedings ICASSP, pp. 5334–5337 (2010)Google Scholar
- 2.Shinoda, T., et al.: “Are you a Werewolf?” becomes a standard problem for general artificial intelligence. In: Proceedings 28th Annual Conference JSAI (2014). (in Japanese)Google Scholar