Detecting Influencers In Very Large Social Networks Of Games

Online games have become a popular form of entertainment, reaching millions of players. Among these players are the game influencers, that is, players with high influence in creating new trends by publishing online content (e.g., videos, blogs, forums). Other players follow the influencers to appreciate their game contents. In this sense, game companies invest in influencers to perform marketing for their products. However, how to identify the game influencers among millions of players of an online game? This paper proposes a framework to extract temporal aspects of the players' actions, and then detect the game influencers by performing a classification analysis. Experiments with the well-known Super Mario Maker game, from Nintendo Inc., Kyoto, Japan, show that our approach is able to detect game influencers of different nations with high accuracy.