DocumentCode :
2824837
Title :
How to deal with bot scum in MMORPGs?
Author :
Hilaire, Sylvain ; Kim, Hyun-chul ; Kim, Chong-Kwon
Author_Institution :
Sch. of Comput. Sci. & Eng., Seoul Nat. Univ., Seoul, South Korea
fYear :
2010
fDate :
8-10 June 2010
Firstpage :
1
Lastpage :
6
Abstract :
Nowadays, bots are becoming a critical issue for the online gaming world. Bots give unfair advantages, and are then considered as cheating and undesirable on game servers. Currently, CAPTCHA and human controls are the most commonly chosen strategies to catch bots. However, these methods are intrusive and complicated have proven to be inefficient due to the large servers´ populations. Researchers have proposed various kinds of automated detection scheme. Yet, these proposals exhibit unpractical features, such as complexity or scalability issues, making the deployment on real systems problematic. We propose to study bots´ and humans´ client-server communication patterns, focusing on a one of most famous MMORPG called World of Warcraft. Intuiting that, for sake of efficiency and human-looking behavior, bots cannot constrain both packet timing and sizes, we propose a detection scheme that combines both parameters. We propose an online algorithm that processes our scheme on the fly as packets arrive. We evaluate the proposed scheme with real packet trace and observe that it can detect bots with small false alarm probability.
Keywords :
client-server systems; computer games; probability; security of data; CAPTCHA strategy; World of Warcraft; automated detection scheme; bot scum; client-server communication patterns; false alarm probability; human control strategy; massively multiplayer online role-playing games; online gaming world; Complexity theory; Correlation; Feature extraction; Games; Humans; Radiation detectors; Servers;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Communications Quality and Reliability (CQR), 2010 IEEE International Workshop Technical Committee on
Conference_Location :
Vancouver, BC
Print_ISBN :
978-1-4244-7795-1
Type :
conf
DOI :
10.1109/CQR.2010.5619911
Filename :
5619911
Link To Document :
بازگشت