Predicting Player Disengagement in Online Games

  • Hanting Xie
  • Daniel Kudenko
  • Sam Devlin
  • Peter Cowling
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 504)


Game engagement, as one of the most fundamental objectives for game designers to achieve, has become an attractive industrial and academic topic. An important direction in this area is to construct a model to predict how long a player could be engaged with a game. This paper introduces a pure data driven method to foresee whether a player will quit the game given their previous activity within the game, by constructing decision trees from historical gameplay data of previous players. The method will be assessed on two popular commercial online games: I Am Playr and Lyroke. The former is a football game while the latter is a music game. The results indicate that the decision tree built by our method is valuable to predict the players’ disengagement and that its human-readable form allow us to search out further reasons about what in-game events made them quit.


Game Data Mining Player Modelling Decision Trees 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    I am playr, (accessed May 23, 2014)
  2. 2.
    Lyroke, (accessed May 23, 2014)
  3. 3.
    We r interactive, (accessed May 23, 2014)
  4. 4.
    Weka, (accessed May 23, 2014)
  5. 5.
    World of warcraft armony, (accessed May 23, 2014)
  6. 6.
    Ahmad, M.A., Keegan, B., Srivastava, J., Williams, D., Contractor, N.: Mining for gold farmers: Automatic detection of deviant players in mmogs. In: International Conference on Computational Science and Engineering, CSE 2009, vol. 4, pp. 340–345. IEEE (2009)Google Scholar
  7. 7.
    Alpaydin, E.: Introduction to machine learning. Adaptive computation and machine learning. MIT Press (2010)Google Scholar
  8. 8.
    Bauckhage, C., Kersting, K., Sifa, R., Thurau, C., Drachen, A., Canossa, A.: How players lose interest in playing a game: An empirical study based on distributions of total playing times. In: 2012 IEEE Conference on Computational Intelligence and Games (CIG), pp. 139–146. IEEE (2012)Google Scholar
  9. 9.
    Bouckaert, R.R., Frank, E., Hall, M., Kirkby, R., Reutemann, P., Seewald, A., Scuse, D.: WEKA Manual for Version 3-7-10. University of Waikato, Hamilton, New Zealand (2013)Google Scholar
  10. 10.
    Drachen, A., Sifa, R., Bauckhage, C., Thurau, C.: Guns, swords and data: Clustering of player behavior in computer games in the wild. In: 2012 IEEE Conference on Computational Intelligence and Games (CIG), pp. 163–170. IEEE (2012)Google Scholar
  11. 11.
    El-Nasr, M., Drachen, A., Canossa, A.: Game Analytics: Maximizing the Value of Player Data. Springer (2013)Google Scholar
  12. 12.
    Hall, M.A.: Correlation-based feature selection for machine learning. Ph.D. thesis, University of Waikato, Hamilton, New Zealand (1999)Google Scholar
  13. 13.
    Mahlmann, T., Drachen, A., Togelius, J., Canossa, A., Yannakakis, G.N.: Predicting player behavior in tomb raider: Underworld. In: 2010 IEEE Conference on Computational Intelligence and Games (CIG), pp. 178–185 (2010)Google Scholar
  14. 14.
    Thurau, C., Bauckhage, C.: Analyzing the evolution of social groups in world of warcraft®. In: 2010 IEEE Conference on Computational Intelligence and Games (CIG), pp. 170–177. IEEE (2010)Google Scholar
  15. 15.
    Weber, B.G., John, M., Mateas, M., Jhala, A.: Modeling Player Retention in Madden NFL 11. In: IAAI (2011)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Hanting Xie
    • 1
  • Daniel Kudenko
    • 1
  • Sam Devlin
    • 1
  • Peter Cowling
    • 1
  1. 1.Department of Computer ScienceUniversity of YorkYorkUK

Personalised recommendations