Portal de Periódicos da CAPES

Time Series and Case-Based Reasoning for an Intelligent Tetris Game

2017; Springer Science+Business Media; Linguagem: Inglês

10.1007/978-3-319-61030-6_13

1611-3349

Diana Sofía Lora Ariza, Antonio A. Sánchez‐Ruiz, Pedro A. González-Calero,

Sports Analytics and Performance

One of the biggest challenges when designing videogames is to keep a player's engagement. Designers try to adapt the game experience for each player defining different difficulty levels or even different sets of behaviors that the non-player characters will use depending on the player profile. It is possible to use different machine learning techniques to automatically classify players in broader groups with distinctive behaviors and then dynamically adjust the game for those types of players. In this paper, we present a case-based approach to detect the skill level of the players in the Tetris game. Cases are extracted from previous game traces and contain time series describing the evolution of a few parameters during the game. Once we know the player level, we adapt the difficulty of the game dynamically providing better or worse Tetris pieces. Our experiments seem to confirm that this type of dynamic difficulty adjustment improves the satisfaction of the players.