A generic method for classification of player behavior

Player classification has allowed us to greatly improve on both game analytics and game adaptivity. With this paper we aim to reverse the ad-hoc tendency in player classification methods. We propose an approach to player classification that could be integrated across different games and genres and is geared towards use by game designers. This paper introduces our generic method of interaction-based player classification. The proposed method consists of three components: (i) intercepting interactions, (ii) finding player types through fuzzy cluster analysis and (iii) classification using Hidden Markov Models. To showcase our method we introduce Blindmaze, a simple web based hidden maze game available to the public, featuring a bounded set of interactions. All data collected from a game is interaction-based, requiring minimal implementation effort from the game’s developers. It is concluded that our method makes player classification even more available by making it generic and re-usable across different games.

@InProceedings { ELB13,
  author       = "Etheredge, Marlon and Lopes, Ricardo and Bidarra, Rafael",
  title        = "A generic method for classification of player behavior",
  booktitle    = "Proceedings of IDPv2 2013 - Workshop on Artificial Intelligence in the Game Design Process, co-located with
                  the Ninth AAAI Conference on Artificial Intelligence in Interactive Digital Entertainment",
  pages        = "2--8",
  month        = "oct",
  year         = "2013",
  publisher    = "AAAI Press",
  organization = "AAAI",
  address      = "AAAI Press, Palo Alto, CA",
  note         = "ISBN 978-1-57735-635-6",
  url          = "http://graphics.tudelft.nl/Publications-new/2013/ELB13"
}