Affective Computation Driven Personalization Modeling in Game-Based Learning

In human-computer interaction and social communication, one's behavior and affective response depend on his personality and environmental stimulus. This means that his behavior and affective response are personalized. In this work, we attempted to build personalization into Game-Based Learning (GBL) system according to user's personality and affective response. User's personality is depicted with a set of attributes such as preference, attitude, character tendency, knowledge & skill, and so on. User's profile is constructed via psychological testing, facial expression recognition, psychophysiology analysis, and behavior analysis. And then, the GBL presents appropriate scenarios and levels adapted to user profile. This will contribute to user's mastery of knowledge & skill, and help user to regulate and improve his personality traits.