Graphical Models as Languages for Computer Assisted Diagnosis and Decision Making

Over the last decade, graphical models for computer assisted diagnosis and decision making have become increasingly popular. Graphical models were originally introduced as ways of decomposing distributions over a large set of variables. However, the main reason for their popularity is that graphs are easy for humans to survey, and most often humans take part in the construction, test, and use of systems for diagnosis and decision making. In other words, at various points in the life cycle of a system, the model is interpreted by a human or communicated between humans. As opposed to machine learning, we shall call this activity human interacted modeling. In this paper we look at graphical models from this point of view. We introduce various kinds of graphical models, and the comprehensibility of their syntax and semantics is in focus.