JoWUA

Abstract

Mobile games are becoming more and more popular nowadays. According to Statista [6], the gaming revenue rises rapidly from 17.6 billion to 40.6 billion from 2013 to 2017. It gradually becomes a big market in the game industry. Game companies are putting more and more focus and resources on their mobile game development. On the other hand, with such a rapid growth of the number of mobile games developed everyday, cheating software developers are also taking advantage of it by implementing and selling cheating software for money. Some famous mobile games (e.g. FIFA Mobile [2], NBA Live Mobile [7], Madden NFL Mobile [5]) include a live market in game for users to trade game items online. This feature is convenient and provides users a unique kind of interactive gaming experience. However, cheaters can take advantage of this feature by implementing auto clickers to automatically buy and bid items in game 24 hours a day effortlessly. In this paper, we are going to investigate this kind of game cheat and propose a detection methodology to determine whether auto clicking behaviour exists. We analyze the touch input behaviour of the client by calculating the clicking dispersion of the touch inputs and the average number of clicks performed per second. Experiment results show that there is a significant difference between the behaviour of auto clickers and human users in term of their clicking positions and frequencies. Our method is easy to implement and the memory consumption is reasonable and practical.