Game Bot Detection Based on Avatar Trajectory
1 like1,260 views
The document discusses a trajectory-based method for detecting game bots in MMORPG and FPS games, addressing challenges such as bots obeying game rules perfectly. It proposes a combination of feature extraction and manifold learning techniques to distinguish between human and bot avatar movements. The study indicates that the approach can be generalized across various game genres and has significant potential for effective bot detection.
Game Bot Detection Based on Avatar Trajectory
- 1. Game Bot Detection Based on Avatar Trajectories Kuan-Ta Chen Academia Sinica Andrew Liao NTU Hsing-Kuo Kenneth Pao NTUST Hao-Hua Chu NTU ICEC IFIP 2008
- 2. Game Bots AI programs that can perform some tasks in place of gamers Popular in MMORPG and FPS games MMORPGs (Massive Multiplayer Role Playing Games) accumulating rewards efficiently in 24 hours a day breaking the balance of power and economies in games FPS games (First-Person Shooting Games) a) improving aiming accuracy b) fully automated achieving high ranking without proficient skills and efforts Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 2
- 3. Bot Detection Detecting whether a character is controlled by a bot is difficult since a bot obeys the game rules perfectly No general detection methods are available today State of practice is identifying via human intelligence Detection: Bots may show regular or peculiar behavior; Confirmation: Bots cannot talk like humans Labor-intensive and may annoy innocent players Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 3
- 4. Earlier Bot Prevention / Detection Work Prevention CAPTCHA tests (Completely Automated Public Turing test to tell Computer and Human Apart) Detection Process monitoring at client side Constantly changing the bot program’s signature Traffic analysis at the network Remove bot traffic’s regularity by heavy-tailed random delays Aiming bot detection using Dynamic Bayesian Network Specific to aiming bots that help aim the target accurately Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 4
- 5. CAPTCHA in a Japanese Online Game Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 5
- 6. Our Goal for Bot Detection Passive detection No intrusion in players’ gaming experience No client software is required Generalizable schemes (for other games and other game genres) Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 6
- 7. Our Solution: Trajectory-based Detection Based on the avatar’s moving trajectory in game Applicable for all genres of games where players control the avatar’s movement directly Avatar’s trajectory is high-dimensional (both in time and spatial domains) Use manifold learning to distinguish the trajectories of human players and game bots Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 7
- 8. The Rationale behind Our Scheme The trajectory of the avatar controlled by a human player is hard to simulate for two reasons: Complex context information: Players control the movement of avatars based on their knowledge, experience, intuition, and a great deal of information provided in the game. Human is not always logical and optimal How to model and simulate realistic movements is still an open question in the AI field?! Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 8
- 9. Bot Detection: A Decision Problem Q: Whether a bot is controlling a game client given the movement trajectory of the avatar? A: Yes / No? Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 9
- 10. Talk Progress Overview Data Description Proposed Schemes Feature-based Manifold learning: Dimensionality Reduction using Isomap Conclusion Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 10
- 11. Case Study: Quake 2 Choose Quake 2 as our case study A classic FPS game Many real-life human traces are available on the Internet more realistic than traces collected in experiments Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 11
- 12. A Screen Shot of Quake 2 Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 12
- 13. Data Collection Human traces downloaded from GotFrag Quake, Planet Quake, Demo Squad, and Revilla Quake Sites Bot traces collected on our own Quake server CR BOT 1.14 Eraser Bot 1.01 ICE Bot 1.0 Each cut into 1,000-second segments Totally 143.8 hours of traces were collected Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 13
- 14. Data Representation t (X, Y) (X, Y) (X, Y) Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 14
- 15. Trails of Human Players Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 15
- 16. Trails of CRBot Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 16
- 17. Trails of Eraser Bot Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 17
- 18. Trails of ICE Bot Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 18
- 19. Talk Progress Overview Data Description Proposed Schemes Feature-based Manifold learning: Dimensionality Reduction using Isomap Conclusion Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 19
- 20. Proposed Schemes Scheme 1 – Feature Based: Extraction of Important Features, Combined with Decision Tree ON/OFF Activity Pace Path Turn Scheme 2 – Manifold Learning Based High-dimensional data Dimension reduction Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 20
- 21. Feature Extraction Given a segment of a trajectory, {xt, yt}, 1 ≤ t ≤ T, we extract the following features from this two-dimensional time series: ON/OFF Activity Pace Path Turn Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 21
- 22. Movement Trail Analysis Activity mean/sd of ON/OFF periods Pace (1 step) speed/offset in each time period teleportation frequency Path (several steps) linger frequency/length smoothness detourness Turn frequency of mild turns, wild turns, and U-turns Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 22
- 23. ON/OFF Activity Features Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 23
- 24. Pace-Related Features Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 24
- 25. Path-Related Features path length detourness = effset Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 25
- 26. Turn-Related Features Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 26
- 27. Data Point Mapping using the First Three Principle Components Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 27
- 28. Classification Results using Decision Tree Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 28
- 29. Proposed Schemes Scheme 1 – Feature Based: Extraction of Important Features, Combined with Decision Tree ON/OFF Activity Pace Path Turn Scheme 2 – Manifold Learning Based High-dimensional data Dimension reduction Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 29
- 30. The Complete Process: Overview Decision Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 30
- 31. Step 1. Pace Vector Construction For trace sn , we compute the pace in consecutive two seconds by sn,i+1 − sn,i = (sn,i+1 − sn,i )T (sn,i+1 − sn,i ) We then compute the distribution of pace lengths with a fixed bin size by Fn = (fn,1 , fn,2 , . . . , fn,B ) where B is the number of bins in the distribution. Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 31
- 32. Examples of Pace Vectors Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 32
- 33. Step 2. Dimension Reduction with Isomap Pace vectors have 201 dimensions We adopt Isomap for dimension reduction for Better performance Reduce computation overhead in classification Isomap Assume data points lie on a smooth manifold Construct the neighborhood graph by kNN (k-nearest neighbors) Compute the shortest path for each pair of points Reconstruct data by MDS (multidimensional scaling) Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 33
- 34. A Graphic Representation of Isomap Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 34
- 35. Distance Measure between Pace Vectors 1. Euclidean Distance 2. Kullback-Leibler divergence (or KL distance) where P & Q are feature vectors of dimension d Because KL distance is not symmetric, we use a symmetric version instead Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 35
- 36. PCA (Linear) vs. Isomap (Nonlinear) Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 36
- 37. Step 3. Classification Apply a supervised classifier on the Isomap-reduced pace vectors To decide whether a pace vector belongs to a bot or a human player Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 37
- 38. Five Methods for Comparison Method Data Input kNN Linear SVM Pace Vectors Nonlinear SVM Isomap + kNN Isomap-reduced Pace Isomap + Nonlinear SVM Vectors Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 38
- 39. Evaluation Results Error Rate False Positive Rate False Negative Rate Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 39
- 40. Talk Progress Overview Data Description Proposed Schemes Feature-based Manifold learning: Dimensionality Reduction using Isomap Conclusion Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 40
- 41. Conclusion We propose a trajectory-based approach for detecting game bots. Feature-based method shows descent result We can improve the result and efficiency further by Isomap + nonlinear SVM approach. Human’s logic in controlling avatars is hard to simulate we believe this approach has the potential to be a general yet robust bot detection methodology Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 41
- 43. Thank You! Hsing-Kuo Kenneth Pao ICEC IFIP 2008
- 44. Addition of Gaussian Noise Bot programmers can try to evade from detection by adding random perturbation into bots’ movement behavior Evaluate the robustness of our scheme by adding Gaussian noise into bots’ trajectories Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 44
- 45. Addition of Gaussian Noise (cont) Human Bots Gaussian Noise Bots with Noise Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 45
- 46. Evaluation Results Error Rate False Positive Rate False Negative Rate Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 46
- 47. Cross-Maps Validation Sometimes human movement may be restricted by the environment around him/her. Whether a classifier trained for a map can be used for detecting bots on another map? The Edge The Frag Pipe Warehouse Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 47
- 48. Evaluation Results Error Rate False Postive Rate False Negative Rate Hsing-Kuo Kenneth Pao / Game Bot Detection Based on Avatar Trajectory 48