Game bot detection using user behavioral characteristics


Abstract

As the online service industry has continued to grow, illegal activities in the online world have drastically increased and become more diverse. Most illegal activities occur continuously because cyber assets, such as game items and cyber money in online games, can be monetized into real currency. The aim of this study is to detect game bots in a massively multiplayer online role playing game (MMORPG). We observed the behavioral characteristics of game bots and found that they execute repetitive tasks associated with gold farming and real money trading. We propose a game bot detection method based on user behavioral characteristics. The method of this paper was applied to real data provided by a major MMORPG company. Detection accuracy rate increased to 96.06 % on the banned account list.


1. Dataset

To perform this study, we rely on a real-world dataset obtained from the operation of Aion, a popular game. Our Game Bot Detection dataset contains all features used in our paper, which were generated from in-game action logs for 88 days, between April 9th and July 5th of 2010. During this period, there were 49,739 characters that played more than 3 h. Among these players, 7702 characters were game bots, identified and labeled by the game company. The banned list was provided by the game company to serve as the ground truth, and each banned user has been vetted and verified by human labor and active monitoring.


1. DataSet Release

For academic purposes, we are happy to release our Dataset. If you use our dataset for your experiment, please cite our paper.

  • Dataset Download Link: Download


The dataset contains the following feature and results.


1.1. Data Processing

Player information features

-> (Data aggregation) -> Results

Player actions features

-> (Frequency & ratio analysis) -> Results

Group activities features

-> (Data aggregation) -> Results

Social interaction diversity features

-> (Entropy analysis) -> Results

Network measures features

-> (Clustering) -> Results

1.2. Feature Selection

  • Best-first Search Weka Result

  • Greedy-stepwise Search Weka Result

  • Information Gain Ranking filter Weka Result

1.3. Classification & Evaluation

    • Decision tree with Feature_Set1 Weka Result

    • Random forest with Feature_Set1 Weka Result

    • Logistic regression with Feature_Set1 Weka Result

    • Naïve Bayes with Feature_Set1 Weka Result

    • Decision tree with Feature_Set2 Weka Result

    • Random forest with Feature_Set2 Weka Result

    • Logistic regression with Feature_Set2 Weka Result

    • Naïve Bayes with Feature_Set2 Weka Result

    • Decision tree with Feature_Set3 Weka Result

    • Random forest with Feature_Set3 Weka Result

    • Logistic regression with Feature_Set3 Weka Result

    • Naïve Bayes with Feature_Set3 Weka Result


2. Publication

Kang, A. R., Jeong, S. H., Mohaisen, A., & Kim, H. K. (2016). Multimodal game bot detection using user behavioral characteristics. SpringerPlus, 5(1), 1-19.


You may want to see the following papers that used our dataset!

[1] M. L. Bernardi, M. Cimitile, F. Martinelli, and F. Mercaldo, “An ensemble fuzzy logic approach to game bot detection through behavioural features,” IEEE Int. Conf. Fuzzy Syst., vol. 2018-July, pp. 1–9, 2018. DOI: 10.1109/FUZZ-IEEE.2018.8491615

[2] M. L. Bernardi, M. Cimitile, F. Martinelli, and F. Mercaldo, “A time series classification approach to game bot detection,” in Proceedings of the 7th International Conference on Web Intelligence, Mining and Semantics - WIMS ’17, 2017, pp. 1–11. DOI: 10.1145/3102254.3102263


3. Contact

          • Ah Reum Kang (armk at hksecurity.net ) or Huy Kang Kim (cenda at korea.ac.kr)