Game bot detection using data mining techniques in the MMORPG

다국어 초록 (Multilingual Abstract)

As the online service industry grows, illegal activities in the online world have drastically increased and become more diverse. Most of the illegal activities are due to the fact that cyber assets such as game items and cyber money in online games can be monetized into real currency. The aim of this thesis is to detect game bots in a Massively Multiplayer Online Role Playing Game (MMORPG). Game activity logs such as the party play and the chat data were used to discriminate abnormal users like game bots from normal users. Through a comprehensive statistical and data mining analysis of user behaviors, the game bot’s behavioral characteristics were investigated. Since these game bots execute repetitive tasks associated with gold-farming and real-money trading, the behavioral differences of game bots and human users were observed. This thesis proposes the game bot detection methodology based on users’ behavioral characteristics. The methodology of this thesis was applied to real data provided by a major MMORPG company. The detection accuracy rate was up to 96.06% on the banned account list.

목차 (Table of Contents)

• 1. Introduction 1
• 2. Party play log analysis 3
• 2.1 Introduction 3
• 2.2 Literature review 5
• 2.3 User behavior analysis for bot detection 13
• 2.4 Results 16
• 2.5 Conclusions 28
• 2.6 Future work 29
• 3. Chat data analysis 30
• 3.1 Introduction 30
• 3.2 Related work 32
• 3.3 Data and text mining framework for game bot detection 36
• 3.4 Experiment results and discussion 43
• 3.5 Conclusion 52
• 4. User’s behavior characteristics analysis 53
• 4.1 Introduction 53
• 4.2 Materials and methods 58
• 4.3 Results and discussion 63
• 4.4 Conclusions 79
• 5. Conclusions 81
• 6. Appendix 82