Clusterization of an Online Game Community through Self-Organizing Maps and an Evolved Fuzzy System

Online games have had an exponential increase in the market. Today, many people interact for hours in a virtual gaming world called the massive multiplayer online role-playing games (MMORPGs). The players maintain relationships and build communities, formed by diverse people who establish links in very different ways. To study the communities formed in those games, it is possible to cluster a player´s community by common characteristics, and examine their relationships. This paper employs self-organizing maps to obtain the clusters of a player community from the game Ragnarok. Aiming to improve the performance of the clusterization, a fuzzy system was designed through genetic algorithms to measure the relevance of the inputs. The results obtained indicate that the proposed methodology can be successfully applied to the clusterization of a multiplayer community.