Probabilistic networks are now fairly well established as practical representations of knowledge for reasoning under uncertainty, as demonstrated by an increasing number of successful applications in such domains as (medical) diagnosis and prognosis, planning, vision,

The assessments obtained for the various conditional probabilities of a Bayesian belief network inevitably are inaccurate. The inaccuracies involved influence the reliability of the network's output. By subjecting the belief network to a sensitivity analysis with respect to its conditional probabilities, the reliability of the output can be investigated. Unfortunately, straightforward sensitivity analysis of a Bayesian belief network is highly time-consuming. In this paper, we show that, by qualitative considerations, several analyses can be identified as being uninformative as the conditional probabilities under study cannot affect the network's output. In addition, we show that the analyses that are informative comply with simple mathematical functions; more specifically, we show that the network's output can be expressed as a quotient of two functions that are linear in a conditional probability under study. These properties allow for considerably reducing the computational burden of se...

elop methods and tools to utilise the temporal data available in a CPR. As statistics lies at the heart of the science of clinical medicine, the project focuses on building temporal Bayesian (probabilistic) models. In particular, we will investigate methods for the exploitation of medical temporal data, methods for expert-guided temporal model development, and learning temporal Bayesian models from temporal data (both structure and parameter learning). In this context, a clinical information system used within an ICU will be used, as such systems are seen as the best approximation to future CPR systems available at the moment. Finally, the clinical information system at UMCU will be used as an environment to investigate the usefulness of the developed methods and tools in a practical, real-life clinical setting. 4 Composition of the Research Team Institute for Computer and Information Sciences, University of Nijmegen (KUN) Toernooiveld 1, 6525 ED Nijmegen, The Netherlands Prof.dr