In this paper, we propose a method for automatically detecting the times during\nwhich game players exhibit specific behavior, such as when players commonly show\nexcitement, concentration, immersion, and surprise. The proposed method detects\nsuch outlying behavior based on the game players� characteristics. These characteristics\nare captured non-invasively in a general game environment. In this paper,\ncameras were used to analyze observed data such as facial expressions and player\nmovements.Moreover, multimodal data from the game players (i.e., data regarding\nadjustments to the volume and the use of the keyboard and mouse) was used to\nanalyze high-dimensional game-player data. A support vector machine was used to\nefficiently detect outlying behaviors. We verified the effectiveness of the proposed\nmethod using games from several genres. The recall rate of the outlying behavior\npre-identified by industry experts was approximately 70%. The proposed method\ncan also be used for feedback analysis of various interactive content provided in PC\nenvironments.
Loading....