A Probabilistic Framework for Recognizing and Affecting Emotions

Architectures for Modeling Emotion: Cross-Disciplinary Foundations

Volume

Papers from the 2004 AAAI Spring Symposium

Contents

We present a framework for affective user modeling that deals with the high level of uncertainty involved in recognizing a variety of user emotions by relying un Dynamic Bayesian Network. We summarize how we used this framework to build a model of player affect to be used by a socially intelligent agent during the interaction with an educational game.