Title :
Hard-core user and bot user classification using game character´s growth types
Author :
Jin Lee; Sung Wook Kang;Huy Kang Kim
Author_Institution :
Graduate School of Information Security, Korea University, Korea
Abstract :
Online game bots unfairly collect items and money, then rapidly deplete the in-game contents. Furthermore, recently game bots have been stealing the gamer´s personal information and cause account thefts problems. These problems can have disastrous effects as well as leading to a downturn in the online gaming industry. There have been various countermeasures to detect game bots. However, misclassification between game bots and hard-core users is the well-known problem for a long time. In this paper, we define the growth types by analyzing the growth processes of users with the Aion dataset, one of the famous MMORPGs in the world. We propose a framework that classifies hard-core users and game bots in the growth patterns. As a result, we successfully distinguish game bots from hard-core users with high accuracy value.
Keywords :
"Games","Feature extraction","Clustering algorithms","Algorithm design and analysis","Multilayer perceptrons","Support vector machines","Information security"
Conference_Titel :
Network and Systems Support for Games (NetGames), 2015 International Workshop on
Electronic_ISBN :
2156-8146
DOI :
10.1109/NetGames.2015.7383000