A Comprehensive Review of LSTM-Based Churn Prediction Models in the Gaming Industry

Document Type : Research Paper

Authors

1 Electrical and Computer Engineering department, Semnan University, Semnan, Iran.

2 Faculty of Electrical and Computer Engineering , Semnan University, Semnan, Iran.

3 Faculty of Electrical and Computer Engineering, Semnan University, Semnan, Iran.

Abstract

Client churn is a significant issue affecting companies across various industries. In the gaming sector, customer loss is particularly critical as it directly impacts revenue, profit margins, and customer retention. Inaccurate predictions of client churn can lead to substantial revenue losses. Churn prediction involves identifying customers who are most likely to cancel their subscriptions. This practice has become essential for many modern organizations due to its performance benefits, aiding businesses in calculating revenue growth and client retention metrics. This paper classifies player churn prediction models into seven main categories to comprehensively review the existing literature. This classification enhances the understanding of various methodologies used in the field and highlights potential areas for future research. Notably, Long Short-Term Memory (LSTM) networks, a type of recurrent neural network, have demonstrated significant potential among deep learning models. This paper examines the contribution of LSTM networks in predicting churn in computer games.

Keywords

Main Subjects

References

[1]           CU. Singh, S. Murugesan, and A. Seth, Emerging Computing Paradigms: Principles, Advances and Applications. Wiley-Blackwell, 2022.
[2]           K. Rothmeier, N. Pflanzl, J. A. Hullmann, and M. Preuss, “Prediction of Player Churn and Disengagement Based on User Activity Data of a Freemium Online Strategy Game,” IEEE Trans. Games, vol. 13, no. 1, pp. 78–88, Mar. 2021, doi: 10.1109/TG.2020.2992282.
[3]           X. Xiahou and Y. Harada, “B2C E-Commerce Customer Churn Prediction Based on K-Means and SVM,” J. Theor. Appl. Electron. Commer. Res., vol. 17, no. 2, pp. 458–475, Apr. 2022, doi: 10.3390/jtaer17020024.
[4]           E. Lee et al., “Game Data Mining Competition on Churn Prediction and Survival Analysis Using Commercial Game Log Data,” IEEE Trans. Games, vol. 11, no. 3, pp. 215–226, Sep. 2019, doi: 10.1109/TG.2018.2888863.
[5]           M. Milošević, N. Živić, and I. Andjelković, “Early churn prediction with personalized targeting in mobile social games,” Expert Syst. Appl., vol. 83, pp. 326–332, 2017, doi: 10.1016/j.eswa.2017.04.056.
[6]           X. Liu et al., “A Semi-Supervised and Inductive Embedding Model for Churn Prediction of Large-Scale Mobile Games,” in 2018 IEEE International Conference on Data Mining (ICDM), IEEE, Nov. 2018, pp. 277–286. doi: 10.1109/ICDM.2018.00043.
[7]           G. Y. Wang, “Churn Prediction for High-Value Players in Freemium Mobile Games: Using Random Under-Sampling,” Statistika, vol. 102, no. 4, pp. 443–453, 2022, doi: 10.54694/STAT.2022.18.
[8]           Y. Hossain, E. Azizi, and L. Zaman, “Predicting subscription renewal using binary classification in World of Warcraft,” Entertain. Comput., vol. 44, no. March 2022, p. 100522, 2025, doi: 10.1016/j.entcom.2022.100522.
[9]           D. David and A. Zahra, “Player Churn Prediction In Free To Play Game Using Ensemble Learning,” Action Res. Lit., vol. 8, no. 4, pp. 543–549, 2024, doi: 10.46799/arl.v8i4.280.
[10]         R. Guo, W. Xiong, Y. Zhang, and Y. Hu, “Enhancing Game Customer Churn Prediction with a Stacked Ensemble Learning Model,” pp. 0–26, 2024.
[11]         A. Zheng, L. Chen, F. Xie, J. Tao, C. Fan, and Z. Zheng, “Keep You from Leaving: Churn Prediction in Online Games,” 2020, pp. 263–279. doi: 10.1007/978-3-030-59416-9_16.
[12]         Z. H. Kilimci, H. Yoruk, and S. Akyokus, “Sentiment Analysis Based Churn Prediction in Mobile Games using Word Embedding Models and Deep Learning Algorithms,” in 2020 International Conference on INnovations in Intelligent SysTems and Applications (INISTA), IEEE, Aug. 2020, pp. 1–7. doi: 10.1109/INISTA49547.2020.9194624.
[13]         D. Halbersberg, R. Yasinnik, M. Halevi, and M. Salhov, “Temporal-Based Action Graph with Sequential Pattern Mining for Churn Detection : a Playtika Case Study,” 2023 Int. Conf. Mach. Learn. Appl., pp. 324–331, 2023, doi: 10.1109/ICMLA58977.2023.00052.
[14]         Y. Xiong et al., “A Data-Driven Decision Support Framework for Player Churn Analysis in Online Games,” Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Min., pp. 5303–5314, 2023, doi: 10.1145/3580305.3599759.
[15]         J. Tao et al., “Explainable AI for Cheating Detection and Churn Prediction in Online Games,” IEEE Trans. Games, vol. 15, no. 2, pp. 242–251, 2023, doi: 10.1109/TG.2022.3173399.
[16]         J. Jeon, D. Yoon, S. Yang, and K. Kim, “Extracting gamers’ cognitive psychological features and improving performance of churn prediction from mobile games,” 2017 IEEE Conf. Comput. Intell. Games, CIG 2017, pp. 150–153, 2017, doi: 10.1109/CIG.2017.8080428.
[17]         S. Kim, D. Choi, E. Lee, and W. Rhee, “Churn prediction of mobile and online casual games using play log data,” PLoS One, vol. 12, no. 7, pp. 1–19, 2017, doi: 10.1371/journal.pone.0180735.
[18]         L. B. Martins Kummer, J. Cesar Nievola, and E. C. Paraiso, “Applying Commitment to Churn and Remaining Players Lifetime Prediction,” in 2018 IEEE Conference on Computational Intelligence and Games (CIG), IEEE, Aug. 2018, pp. 1–8. doi: 10.1109/CIG.2018.8490443.
[19]         J. T. Kristensen and P. Burelli, “Combining Sequential and Aggregated Data for Churn Prediction in Casual Freemium Games,” in 2019 IEEE Conference on Games (CoG), IEEE, Aug. 2019, pp. 1–8. doi: 10.1109/CIG.2019.8848106.
[20]         A. Perisic and M. Pahor, “Extended RFM logit model for churn prediction in the mobile gaming market,” Croat. Oper. Res. Rev., vol. 11, no. 2, pp. 249–261, 2020, doi: 10.17535/crorr.2020.0020.
[21]         A. Perisic and M. Pahor, “RFM-LIR Feature Framework for Churn Prediction in the Mobile Games Market,” IEEE Trans. Games, vol. 14, no. 2, pp. 126–137, Jun. 2022, doi: 10.1109/TG.2021.3067114.
[22]         W. Yang et al., “Mining Player In-game Time Spending Regularity for Churn Prediction in Free Online Games,” in 2019 IEEE Conference on Games (CoG), IEEE, Aug. 2019, pp. 1–8. doi: 10.1109/CIG.2019.8848033.
[23]         W. Yang, T. Huang, J. Zeng, L. Chen, S. Mishra, and Y. E. Liu, “Utilizing Players’ Playtime Records for Churn Prediction: Mining Playtime Regularity,” IEEE Trans. Games, vol. 14, no. 2, pp. 153–160, Jun. 2022, doi: 10.1109/TG.2020.3024829.
[24]         M. Óskarsdóttir, K. E. Gísladóttir, R. Stefánsson, D. Aleman, and C. Sarraute, “Social networks for enhanced player churn prediction in mobile free-to-play games,” Appl. Netw. Sci., vol. 7, no. 1, 2022, doi: 10.1007/s41109-022-00524-5.
[25]         Y. Han and J. Moon, “Prediction of Churning Game Users Based on Social Activity and Churn Graph Neural Networks,” IEEE Access, vol. 12, no. June, pp. 101971–101984, 2024, doi: 10.1109/ACCESS.2024.3429559.
[26]         A. Perianez, A. Saas, A. Guitart, and C. Magne, “Churn Prediction in Mobile Social Games : Towards a Complete Assessment Using Survival Ensembles”.
[27]         P. Bertens and A. Guitart, “Games and Big Data : A Scalable Multi-Dimensional Churn Prediction Model”.
[28]         S. Demediuk et al., “Player retention in League of Legends,” in Proceedings of the Australasian Computer Science Week Multiconference, New York, NY, USA: ACM, Jan. 2018, pp. 1–9. doi: 10.1145/3167918.3167937.
[29]         V. Bonometti, C. Ringer, M. Hall, A. R. Wade, and A. Drachen, “Modelling Early User-Game Interactions for Joint Estimation of Survival Time and Churn Probability,” in 2019 IEEE Conference on Games (CoG), IEEE, Aug. 2019, pp. 1–8. doi: 10.1109/CIG.2019.8848038.
[30]         K. Jang, J. Kim, and B. Yu, “On Analyzing Churn Prediction in Mobile Games,” in 2021 6th International Conference on Machine Learning Technologies, New York, NY, USA: ACM, Apr. 2021, pp. 20–25. doi: 10.1145/3468891.3468895.
[31]         K. Jang, J. Kim, and B. Yu, “Vector-based Churn Prediction using Neural Networks in Mobile Games,” in 2019 IEEE International Conference on Big Data (Big Data), IEEE, Dec. 2019, pp. 6081–6083. doi: 10.1109/BigData47090.2019.9006052.
[32]         S. Roohi, A. Relas, J. Takatalo, H. Heiskanen, and P. Hämäläinen, “Predicting Game Difficulty and Churn Without Players,” Aug. 2020, doi: 10.1145/3410404.3414235.
[33]         J. Li et al., A Difficulty-Aware Framework for Churn Prediction and Intervention in Games, vol. 1, no. 1. Association for Computing Machinery, 2021. doi: 10.1145/3447548.3467277.
[34]         A. F. del Río, A. Guitart, and Á. Periánez, “A time series approach to player churn and conversion in videogames,” Intell. Data Anal., vol. 25, no. 1, pp. 177–203, 2021, doi: 10.3233/IDA-194940.
[35]         B. Karmakar and P. Liu, “Improved retention analysis in freemium playing games by jointly modelling players ’ motivation, progression and churn,” no. July 2020, pp. 1–32, 2021, doi: 10.1111/rssa.12730.
[36]         A. Guitart, P. Chen, and Á. Periáñez, “The Winning Solution to the IEEE CIG 2017 Game Data Mining Competition,” Mach. Learn. Knowl. Extr., vol. 1, no. 1, pp. 252–264, Dec. 2018, doi: 10.3390/make1010016.
[37]         M. Christ, N. Braun, J. Neuffer, and A. W. Kempa-liehr, “Neurocomputing Time Series FeatuRe Extraction on basis of Scalable Hypothesis tests ( tsfresh – A Python package ),” Neurocomputing, vol. 307, pp. 72–77, 2018, doi: 10.1016/j.neucom.2018.03.067.
Modeling and Simulation in Electrical and Electronics Engineering
Volume 3, Issue 3 - Serial Number 13
November 2023
Pages 51-60

Files

History

  • Receive Date: 24 November 2023
  • Revise Date: 13 August 2024
  • Accept Date: 06 January 2025

Share

How to cite

Statistics