한국게임학회 논문지 (Journal of Korea Game Society)

  • 제22권1호
  • /
  • Pages.33-42
  • /
  • 2022
  • /
  • 1598-4540(pISSN)
  • /
  • 2287-8211(eISSN)
한국게임학회 (Korea Game Society)
권호 표지
DOI QR코드

DOI QR Code

기계학습을 활용한 온라인게임 매치메이킹 개선방안

Improvement of online game matchmaking using machine learning

  • 김용우 (한양대학교 기술경영학과) ;
  • 김영민 (한양대학교 기술경영학과)
  • Kim, Yongwoo (Department of Technology Management, Hanyang University) ;
  • Kim, Young‐Min (Department of Technology Management, Hanyang University)
  • 투고 : 2021.11.10
  • 심사 : 2021.12.23
  • 발행 : 2022.02.20
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

온라인 게임에서 다른 플레이어와의 상호작용은 플레이어의 만족도에 영향을 미친다. 따라서, 비슷한 수준의 플레이어를 매치시켜 원활한 상호작용을 도모하는 것은 플레이어의 게임 경험을 위해 중요하다. 그러나, 게임의 최종승패로만 플레이어의 평가점수를 증감시키는 현재의 평가 방식으로는 신규 및 복귀 플레이어의 원활한 매칭이 불가능하다. 본 연구에서는 스타크래프트II의 리플레이를 활용하여 매치메이킹 개선을 위한 기계학습 활용방안을 제시한다. 매치메이킹의 기준이 되는 플레이어의 MMR 점수를 예측하는 기계학습 모델을 생성하고 성능을 평가하였다. 모델의 오차는 리그 평균 MMR 점수 범위의 40.4% 수준으로, 제안된 방식을 통해서 플레이어를 실력과 근접한 리그에 즉시 배치할 수 있음을 확인하였다. 또한, 결과에 대한 플레이어의 수용도를 높일 수 있도록 예측의 근거를 도출하는 방안도 제시되었다.

In online games, interactions with other players may threaten player satisfaction. Therefore, matching players of similar skill levels is important for players' experience. However, with the current evaluation method which is only based on the final result of the game, newbies and returning players are difficult to be matched properly. In this study, we propose a method to improve matchmaking quality. We build machine learning models to predict the MMR of players and derive the basis of the prediction. The error of the best model was 40.4% of the average MMR range, confirming that the proposed method can immediately place players in a league close to their current skill level. In addition, the basis of predictions may help players to accept the result.

키워드

참고문헌

  1. Bateman, C., & Boon, R., "21st Century Game Design (Game Development Series)", Charles River Media, Inc., 2005.
  2. Charles, D., McNeill, M., McAlister, M., Black, M., Moore, A., Stringer, K., ... & Kerr, A., "Player-centred game design: Player modelling and adaptive digital games", Proceedings of DiGRA 2005 Conference, pp. 285-298, 2005.
  3. Lazzaro, N., "Why we play games: Four keys to more emotion in player experiences", Proceedings of GDC, Vol. 306, pp. 1-8, 2004.
  4. Google LLC., "Open Match", accessed Dec 14. 2021, https://development.open-match.dev/.
  5. Elo, Arpad E., "The rating of chessplayers, past and present", Arco Pub., 1978.
  6. Glickman, Mark E., "Parameter estimation in large dynamic paired comparison experiments", Applied Statistics, Vol. 48, No. 3, pp. 377-394, 1999. https://doi.org/10.1111/1467-9876.00159
  7. Glickman, Mark E., "Dynamic paired comparison models with stochastic variances", Journal of Applied Statistics, Vol. 28, No. 6, pp. 673-689, 2001. https://doi.org/10.1080/02664760120059219
  8. Herbrich, R., Minka, T. & Graepel, T., "TrueSkill™: a Bayesian skill rating system", Proceedings of the 19th International Conference on Neural Information Processing Systems, pp. 569-576, 2006.
  9. Robertson, G., & Watson, I., "An improved dataset and extraction process for starcraft ai." The Twenty-Seventh International Flairs Conference, 2014.
  10. Synnaeve, G., & Bessiere, P., "A dataset for StarCraft AI and an example of armies clustering", Eighth Artificial Intelligence and Interactive Digital Entertainment Conference, 2012.
  11. Lin, Z., Gehring, J., Khalidov, V., and Synnaeve, G., "Stardata: A starcraft ai research dataset", Thirteenth Artificial Intelligence and Interactive Digital Entertainment Conference, 2017.
  12. Ravari, Y. N., Bakkes, S., & Spronck, P., "Starcraft winner prediction", Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment, Vol. 12. No. 1, 2016.
  13. Stanescu, M., Hernandez, S. P., Erickson, G., Greiner, R., & Buro, M., "Predicting army combat outcomes in StarCraft", Ninth artificial intelligence and interactive digital entertainment conference, 2013.
  14. Hsieh, J. L., & Sun, C. T., "Building a player strategy model by analyzing replays of real-time strategy games", 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence), pp. 3106-3111, 2008.
  15. Weber, B. G., & Mateas, M., "A data mining approach to strategy prediction", 2009 IEEE Symposium on Computational Intelligence and Games, pp. 140-147, 2009.
  16. Schadd, F., Bakkes, S., & Spronck, P., "Opponent Modeling in Real-Time Strategy Games", Proceedings of the GAME-ON, pp. 61-70, 2007.
  17. Vinyals, O., Babuschkin, I., Czarnecki, W. M., Mathieu, M., Dudzik, A., Chung, J., ... & Silver, D., "Grandmaster level in StarCraft II using multi-agent reinforcement learning", Nature, Vol. 575, No. 7782, pp. 350-354, 2019. https://doi.org/10.1038/s41586-019-1724-z
  18. Thompson JJ, Blair MR, Chen L & Henrey AJ, "Video Game Telemetry as a Critical Tool in the Study of Complex Skill Learning", PLoS ONE 8(9): e75129, 2013. https://doi.org/10.1371/journal.pone.0075129
  19. Avontuur, T., Spronck, P., & Van Zaanen, M., "Player skill modeling in Starcraft II", Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment, Vol. 9, No. 1, 2013.
  20. Lee, C. M., & Ahn, C. W., "Feature Extraction for StarCraft II League Prediction." Electronics, Vol. 10, No. 8, 2021.
  21. Wolpert, D. H., "Stacked generalization", Neural networks, Vol. 5, No. 2, pp. 241-259, 1992. https://doi.org/10.1016/S0893-6080(05)80023-1
  22. Breiman, L, "Stacked regressions", Machine learning, Vol. 24, No. 1, pp. 49-64, 1996. https://doi.org/10.1007/BF00117832
  23. van der Laan, M., Polley, E. & Hubbard, A., "Super Learner", Statistical Applications in Genetics and Molecular Biology, Vol. 6, No. 1, 2007.
  24. Ribeiro, M. T., Singh, S., & Guestrin, C., ""Why should i trust you?" Explaining the predictions of any classifier", Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining, pp. 1135-1144, 2016.
  25. Lundberg, S. M., & Lee, S. I., "A unified approach to interpreting model predictions", Proceedings of the 31st international conference on neural information processing systems, pp. 4768-4777, 2017.
  26. Sundararajan, M., & Najmi, A., "The many Shapley values for model explanation", International Conference on Machine Learning, pp. 9269-9278, 2020.
  27. Garcia, M. V., & Aznarte, J. L., "Shapley additive explanations for NO2 forecasting", Ecological Informatics, Vol. 56, No. 101039, 2020.
  28. Roth, A. E. (Ed.) , "The Shapley value: essays in honor of Lloyd S Shapley", Cambridge University Press, 1988.
  29. Hegtvedt, K. A., & Killian, C., "Fairness and Emotions: Reactions to the Process and Outcomes of Negotiations", Social Forces, Vol. 78, No. 1, pp. 269-302, 1999. https://doi.org/10.1093/sf/78.1.269
  30. Maute, M. F., & Dubes, L., "Patterns of emotional responses and behavioural consequences of dissatisfaction", Applied Psychology, Vol. 48, No. 3, pp. 349-366, 1999. https://doi.org/10.1111/j.1464-0597.1999.tb00006.x