User Adaptive Game Characters Using Decision Trees and FSMs

Recently, various ways are being explored for enhancing the fun of computer games and lengthening the life cycle of them. Some games, add realistic graphic effect and excellent acoustic effect, and make the tendencies of game players reflected. This paper suggests the method to collect and analyze the action patterns of game players. The game players' patterns are modeled using FSM (Finite State Machine). The result obtained by analyzing the data on game players is used for creating game-agents which show new action patterns by altering the FSM defined previously. This characters are adaptable game-agent which is learnable the action patterns of game players. The proposal method can be applied to create characters which play the role of partners with game players or the role of enemies against game players.