Churn prediction in MMORPGs using player motivation theories and an ensemble approach

Zoheb Borbora, Jaideep Srivastava, Kuo Wei Hsu, Dmitri Wil Iams

Research output: Chapter in Book/Report/Conference proceedingConference contribution

25 Citations (Scopus)

Abstract

In this paper, we investigate the problem of churn prediction in Massively multiplayer online role-playing games (MMORPGs) from a social science perspective and develop models incorporating theories of player motivation. The ability to predict player churn can be a valuable resource to game developers designing customer retention strategies. The results from our theory-driven model significantly outperform a diffusion-based churn prediction model on the same dataset. We describe the synthesis between a theory-driven approach and a data-driven approach to a problem and examine the trade-offs involved between the two approaches in terms of prediction accuracy, interpretability and model complexity. We observe that even though the theory-driven model is not as accurate as the data-driven one, the theory-driven model itself can be more interpretable to the domain experts and hence, more preferable over a complex data-driven model. We perform lift analysis of the two models and find that if a marketing effort is restricted in the number of customers it can contact, the theory-driven model would offer much better return-on-investment by identifying more customers among that restricted set who have the highest probability of churn. Finally, we use a clustering technique to partition the dataset and then build an ensemble on the partitioned dataset for better performance. Experiment results show that the ensemble performs notably better than the single classifier in terms of its recall value, which is a highly desirable property in the churn prediction problem.

Original languageEnglish
Title of host publicationProceedings - 2011 IEEE International Conference on Privacy, Security, Risk and Trust and IEEE International Conference on Social Computing, PASSAT/SocialCom 2011
Pages157-164
Number of pages8
DOIs
Publication statusPublished - 2011
Externally publishedYes
Event2011 IEEE International Conference on Privacy, Security, Risk and Trust, PASSAT 2011 and 2011 IEEE International Conference on Social Computing, SocialCom 2011 - Boston, MA
Duration: 9 Oct 201111 Oct 2011

Other

Other2011 IEEE International Conference on Privacy, Security, Risk and Trust, PASSAT 2011 and 2011 IEEE International Conference on Social Computing, SocialCom 2011
CityBoston, MA
Period9/10/1111/10/11

Fingerprint

Social sciences
Marketing
Classifiers
Experiments

ASJC Scopus subject areas

  • Hardware and Architecture
  • Safety, Risk, Reliability and Quality

Cite this

Borbora, Z., Srivastava, J., Hsu, K. W., & Iams, D. W. (2011). Churn prediction in MMORPGs using player motivation theories and an ensemble approach. In Proceedings - 2011 IEEE International Conference on Privacy, Security, Risk and Trust and IEEE International Conference on Social Computing, PASSAT/SocialCom 2011 (pp. 157-164). [6113108] https://doi.org/10.1109/PASSAT/SocialCom.2011.122

Churn prediction in MMORPGs using player motivation theories and an ensemble approach. / Borbora, Zoheb; Srivastava, Jaideep; Hsu, Kuo Wei; Iams, Dmitri Wil.

Proceedings - 2011 IEEE International Conference on Privacy, Security, Risk and Trust and IEEE International Conference on Social Computing, PASSAT/SocialCom 2011. 2011. p. 157-164 6113108.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Borbora, Z, Srivastava, J, Hsu, KW & Iams, DW 2011, Churn prediction in MMORPGs using player motivation theories and an ensemble approach. in Proceedings - 2011 IEEE International Conference on Privacy, Security, Risk and Trust and IEEE International Conference on Social Computing, PASSAT/SocialCom 2011., 6113108, pp. 157-164, 2011 IEEE International Conference on Privacy, Security, Risk and Trust, PASSAT 2011 and 2011 IEEE International Conference on Social Computing, SocialCom 2011, Boston, MA, 9/10/11. https://doi.org/10.1109/PASSAT/SocialCom.2011.122
Borbora Z, Srivastava J, Hsu KW, Iams DW. Churn prediction in MMORPGs using player motivation theories and an ensemble approach. In Proceedings - 2011 IEEE International Conference on Privacy, Security, Risk and Trust and IEEE International Conference on Social Computing, PASSAT/SocialCom 2011. 2011. p. 157-164. 6113108 https://doi.org/10.1109/PASSAT/SocialCom.2011.122
Borbora, Zoheb ; Srivastava, Jaideep ; Hsu, Kuo Wei ; Iams, Dmitri Wil. / Churn prediction in MMORPGs using player motivation theories and an ensemble approach. Proceedings - 2011 IEEE International Conference on Privacy, Security, Risk and Trust and IEEE International Conference on Social Computing, PASSAT/SocialCom 2011. 2011. pp. 157-164
@inproceedings{a72863a3e175442bad28fd00cc44cdbe,
title = "Churn prediction in MMORPGs using player motivation theories and an ensemble approach",
abstract = "In this paper, we investigate the problem of churn prediction in Massively multiplayer online role-playing games (MMORPGs) from a social science perspective and develop models incorporating theories of player motivation. The ability to predict player churn can be a valuable resource to game developers designing customer retention strategies. The results from our theory-driven model significantly outperform a diffusion-based churn prediction model on the same dataset. We describe the synthesis between a theory-driven approach and a data-driven approach to a problem and examine the trade-offs involved between the two approaches in terms of prediction accuracy, interpretability and model complexity. We observe that even though the theory-driven model is not as accurate as the data-driven one, the theory-driven model itself can be more interpretable to the domain experts and hence, more preferable over a complex data-driven model. We perform lift analysis of the two models and find that if a marketing effort is restricted in the number of customers it can contact, the theory-driven model would offer much better return-on-investment by identifying more customers among that restricted set who have the highest probability of churn. Finally, we use a clustering technique to partition the dataset and then build an ensemble on the partitioned dataset for better performance. Experiment results show that the ensemble performs notably better than the single classifier in terms of its recall value, which is a highly desirable property in the churn prediction problem.",
author = "Zoheb Borbora and Jaideep Srivastava and Hsu, {Kuo Wei} and Iams, {Dmitri Wil}",
year = "2011",
doi = "10.1109/PASSAT/SocialCom.2011.122",
language = "English",
isbn = "9780769545783",
pages = "157--164",
booktitle = "Proceedings - 2011 IEEE International Conference on Privacy, Security, Risk and Trust and IEEE International Conference on Social Computing, PASSAT/SocialCom 2011",

}

TY - GEN

T1 - Churn prediction in MMORPGs using player motivation theories and an ensemble approach

AU - Borbora, Zoheb

AU - Srivastava, Jaideep

AU - Hsu, Kuo Wei

AU - Iams, Dmitri Wil

PY - 2011

Y1 - 2011

N2 - In this paper, we investigate the problem of churn prediction in Massively multiplayer online role-playing games (MMORPGs) from a social science perspective and develop models incorporating theories of player motivation. The ability to predict player churn can be a valuable resource to game developers designing customer retention strategies. The results from our theory-driven model significantly outperform a diffusion-based churn prediction model on the same dataset. We describe the synthesis between a theory-driven approach and a data-driven approach to a problem and examine the trade-offs involved between the two approaches in terms of prediction accuracy, interpretability and model complexity. We observe that even though the theory-driven model is not as accurate as the data-driven one, the theory-driven model itself can be more interpretable to the domain experts and hence, more preferable over a complex data-driven model. We perform lift analysis of the two models and find that if a marketing effort is restricted in the number of customers it can contact, the theory-driven model would offer much better return-on-investment by identifying more customers among that restricted set who have the highest probability of churn. Finally, we use a clustering technique to partition the dataset and then build an ensemble on the partitioned dataset for better performance. Experiment results show that the ensemble performs notably better than the single classifier in terms of its recall value, which is a highly desirable property in the churn prediction problem.

AB - In this paper, we investigate the problem of churn prediction in Massively multiplayer online role-playing games (MMORPGs) from a social science perspective and develop models incorporating theories of player motivation. The ability to predict player churn can be a valuable resource to game developers designing customer retention strategies. The results from our theory-driven model significantly outperform a diffusion-based churn prediction model on the same dataset. We describe the synthesis between a theory-driven approach and a data-driven approach to a problem and examine the trade-offs involved between the two approaches in terms of prediction accuracy, interpretability and model complexity. We observe that even though the theory-driven model is not as accurate as the data-driven one, the theory-driven model itself can be more interpretable to the domain experts and hence, more preferable over a complex data-driven model. We perform lift analysis of the two models and find that if a marketing effort is restricted in the number of customers it can contact, the theory-driven model would offer much better return-on-investment by identifying more customers among that restricted set who have the highest probability of churn. Finally, we use a clustering technique to partition the dataset and then build an ensemble on the partitioned dataset for better performance. Experiment results show that the ensemble performs notably better than the single classifier in terms of its recall value, which is a highly desirable property in the churn prediction problem.

UR - http://www.scopus.com/inward/record.url?scp=84856202925&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84856202925&partnerID=8YFLogxK

U2 - 10.1109/PASSAT/SocialCom.2011.122

DO - 10.1109/PASSAT/SocialCom.2011.122

M3 - Conference contribution

AN - SCOPUS:84856202925

SN - 9780769545783

SP - 157

EP - 164

BT - Proceedings - 2011 IEEE International Conference on Privacy, Security, Risk and Trust and IEEE International Conference on Social Computing, PASSAT/SocialCom 2011

ER -