Mobile Intelligent Synthetic Character in Smartphone Using Behavior Selection Network

Since smart phones have become popular, intelligent adaptive services were more highly demanded. The intelligent synthetic character on a mobile phone is one of the promising services. It is difficult to develop an intelligent synthetic character on the mobile environment due to its dynamism and complexity. This paper proposes a method to generate a synthetic character´s behaviors using contexts inferred with Bayesian networks. To make more realistic character, the OCC model creates the character´s emotion. Large-scale modular behavior networks generate the character´s behaviors with inferred contexts. In order to verify the naturalness and flexibility of the generated behaviors, a simulation is conducted with the mobile log collected with a Samsung SPH-M4650 smart phone.