Ensemble learning for detecting game bot in MMORPG†
Hyunsue Jung1 · Jungwon Choi2 · Sangjun Weon3 · Suin Kim4 · Jongbeom Park5 · Nayeon Lee6 · Yoonsuh Jung7
1LG CNS
2347Department of Statistics, Korea University
56Kakao Games Corp.
2347Department of Statistics, Korea University
56Kakao Games Corp.
Correspondence to: † Jung’s work has been partially supported by National Research Foundation of Korea (NRF) grants funded by the Korean government (MIST) (No. 2022R1F1A1071126 and No. 2022M3J6A1063595).
1 Enterprise Analytics 1 Team, LG CNS, 71, Magokjungang 8-ro, Gangseo-gu, Seoul 07795, Korea.
2 Graduate Student, Department of Statistics, Korea University, Seoul 02841, Korea
3 Graduate Student, Department of Statistics, Korea University, Seoul 02841, Korea
4 Graduate Student, Department of Statistics, Korea University, Seoul 02841, Korea
5 Data Analytics Labs, Kakao Games Corp., Gyeonggi 13529, Korea.
6 Data Analytics Labs, Kakao Games Corp., Gyeonggi 13529, Korea.
7 Professor, Department of Statistics, Korea University, Seoul 02841, Korea. Email: yoons77@korea.ac.kr
1 Enterprise Analytics 1 Team, LG CNS, 71, Magokjungang 8-ro, Gangseo-gu, Seoul 07795, Korea.
2 Graduate Student, Department of Statistics, Korea University, Seoul 02841, Korea
3 Graduate Student, Department of Statistics, Korea University, Seoul 02841, Korea
4 Graduate Student, Department of Statistics, Korea University, Seoul 02841, Korea
5 Data Analytics Labs, Kakao Games Corp., Gyeonggi 13529, Korea.
6 Data Analytics Labs, Kakao Games Corp., Gyeonggi 13529, Korea.
7 Professor, Department of Statistics, Korea University, Seoul 02841, Korea. Email: yoons77@korea.ac.kr
Received April 18, 2024; Revised July 17, 2024; Accepted July 24, 2024.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
- Abstract
- As massively multiplayer online role-playing games (MMORPG) grow in popularity, illegal groups, called gold farmer groups, that operate automated game bots to collect in-game goods for cash gain are becoming a more serious problem. These groups are disrupting the in-game marketplace by increasing stress for normal users. Therefore, making proper bot detection is a critical challenge for game managers. The existing bot detection methods are mostly in-game behavioral analysis and rule-based systems. However, these methods has a limitation because game bots are constantly changing to mimic the game behavior of normal users. In this study, we propose a new bot detection model with ensemble learning. Based on the idea that the objective of behaviors of game bots is gaining in-game goods, their cash flow patterns and item transaction patterns does not change. Therefore, we suggest to use the variables that are related to abnormal item transactions and cash flow patterns. Our proposed model is an ensemble of deep learning models and machine learning (ML) models.
- Keywords : Ensemble model, game bot detection, LSTM, MMORPG
Full Text(PDF) Free
