Investigation of game players’ strategies using visualization techniques and data mining

Nowadays, generating a profitable game is a big challenge. Thousands of commercial products are generated annually that compete with each other for getting more market share. Many tools and techniques, such as project management and user testing are used for effective game development. In recent years, game developers take notice to gamers’ behavior and tend to design and modify games based on their attitude. In this paper, we present an approach called UPB, for analyzing computer games based on telemetry data, statistical information, and data mining. The goal is developing games in a way that improves user experience and increases player enjoyment, so more gamers would welcome them. For this purpose, a telemetry data gathering system and a software for analyzing telemetry data were developed. A data mining process was also conducted to discover the behavior patterns of the game players based on the telemetry data collected from them. For this purpose, we designed a new game, called “Treasure Castle”, to show the effectiveness of UPB approach and data gathering. The difficulty of game levels was determined based on the behavior of the players. In addition, the playing style of players were visualized and their strategies were detected by examining how they play. Our experiments were done on the Treasure Castle game and it has been shown that the proposed approach can answer different questions about understanding players’ behavior.