We present an approach to the solution of de cision problems formulated as influence dia grams. This approach involves a special tri angulation of the underlying graph, the con struction of a junction tree with special prop erties, and a message passing algorithm op erating on the junction tree for computa tion of expected utilities and optimal decision policies.
Causal probabilistic networks have proved to be a useful knowledge representation tool for domains having a natural description in terms of causal relations involving uncertainty between domain concepts. This article describes a network modeling diseases affecting the median nerve. The qualitative structure of the model and the quantitative pathophysiological MUNIN stands for MUscle and Nerve Inference Network. According to Norse mythology MUNIN is a raven whispering intelligence to the god Odin. 301 [385] Downloaded by [University of Sussex Library] at 19:33 03 February 2015 302 [386J K. G. Olesen et el.models were derived from textbooks, statistical material, and clinical experience. The resulting causal probabilistic network is not a tree; it contains numerous loops. A new inference method capable of handling loops was implemented. Two methods were used to minimize the network's requirements for memory space and computation time: "divorcing of multiple parents" and "close-to-optimal" triangulation of the network. The network can be used as a simulation tool, that is, to generate a set of expected findings for different diseases. It can also be used as a diagnostic tool, to provide probabilities of disease states, given a set of findings. In both cases the network performs medically satisfactorily.
As the framework of probabilistic graphical models becomes increasingly popular for knowledge representation and inference, the need for efficient tools for its support is increasing. The Hugin Tool is a general purpose tool for construction, maintenance, and deployment of Bayesian networks and influence diagrams. This paper surveys the key functionality of the Hugin Tool and reports on new advances of the tool. Furthermore, an empirical analysis reports on the efficiency of the Hugin Tool on common inference and learning tasks.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.