Three paediatric cardiologists assessed nearly 1000 imprecise subjective conditional probabilities for a simple belief network representing congenital heart disease, and the quality of the assessments has been measured using prospective data on 200 babies. Quality has been assessed by a Brier scoring rule, which decomposes into terms measuring lack of discrimination and reliability. The results are displayed for each of 27 diseases and 24 questions, and generally the assessments are reliable although there was a tendency for the probabilities to be too extreme. The imprecision allows the judgements to be converted to implicit samples, and by combining with the observed data the probabilities naturally adapt with experience. This appears to be a practical procedure even for reasonably large expert systems.

This paper discusses a new measure that is adaptable to certain intervalic probability frameworks, possibility theory, and belief theory. As such, it has the potential for wide use in knowledge engineering, expert systems, and related problems in the human sciences. This measure (denoted here by F) has been introduced in Smithson (1988) and is more formally discussed in Smithson (1989a)o Here, I propose to outline the conceptual basis for F and compare its properties with other measures of second-order uncertainty. I will argue that F is an indicator of nonspecificity or alternatively, of freedom, as distinguished from either ambiguity or vagueness.

During the summer of 1989, the Forecast Systems Laboratory of the National Oceanic and Atmospheric Administration sponsored an evaluation of artificial intelligence-based systems that forecast severe convective storms. The evaluation experiment, called Shootout-89, took place in Boulder, and focussed on storms over the northeastern Colorado foothills and plains (Moninger, et al., 1990). Six systems participated in Shootout-89. These included traditional expert systems, an analogy-based system, and a system developed using methods from the cognitive science/judgment analysis tradition. Each day of the exercise, the systems generated 2 to 9 hour forecasts of the probabilities of occurrence of: non significant weather, significant weather, and severe weather, in each of four regions in northeastern Colorado. A verification coordinator working at the Denver Weather Service Forecast Office gathered ground-truth data from a network of observers. Systems were evaluated on the basis of several measures of forecast skill, and on other metrics such as timeliness, ease of learning, and ease of use. Systems were generally easy to operate, however the various systems required substantially different levels of meteorological expertise on the part of their users--reflecting the various operational environments for which the systems had been designed. Systems varied in their statistical behavior, but on this difficult forecast problem, the systems generally showed a skill approximately equal to that of persistence forecasts and climatological (historical frequency) forecasts. The two systems that appeared best able to discriminate significant from non significant weather events were traditional expert systems. Both of these systems required the operator to make relatively sophisticated meteorological judgments. We are unable, based on only one summer's worth of data, to determine the extent to which the greater skill of the two systems was due to the content of their knowledge bases, or to the subjective judgments of the operator. A follow-on experiment, Shootout-91, is currently being planned. Interested potential participants are encouraged to contact the author at the address above.

Many Artificial Intelligence systems depend on the agent's updating its beliefs about the world on the basis of experience. Experiments constitute one type of experience, so scientific methodology offers a natural environment for examining the issues attendant to using this class of evidence. This paper presents a framework which structures the process of using scientific data from research reports for the purpose of making decisions, using decision analysis as the basis for the structure and using medical research as the general scientific domain. The structure extends the basic influence diagram for updating belief in an object domain parameter of interest by expanding the parameter into four parts: those of the patient, the population, the study sample, and the effective study sample. The structure uses biases to perform the transformation of one parameter into another, so that, for instance, selection biases, in concert with the population parameter, yield the study sample parameter. The influence diagram structure provides decision theoretic justification for practices of good clinical research such as randomized assignment and blindfolding of care providers. The model covers most research designs used in medicine: case-control studies, cohort studies, and controlled clinical trials, and provides an architecture to separate clearly between statistical knowledge and domain knowledge. The proposed general model can be the basis for clinical epidemiological advisory systems, when coupled with heuristic pruning of irrelevant biases; of statistical workstations, when the computational machinery for calculation of posterior distributions is added; and of meta-analytic reviews, when multiple studies may impact on a single population parameter.

We examine a probabilistic model for the diagnosis of multiple diseases. In the model, diseases and findings are represented as binary variables. Also, diseases are marginally independent, features are conditionally independent given disease instances, and diseases interact to produce findings via a noisy OR-gate. An algorithm for computing the posterior probability of each disease, given a set of observed findings, called quickscore, is presented. The time complexity of the algorithm is O(nm-2m+), where n is the number of diseases, m+ is the number of positive findings and m- is the number of negative findings. Although the time complexity of quickscore i5 exponential in the number of positive findings, the algorithm is useful in practice because the number of observed positive findings is usually far less than the number of diseases under consideration. Performance results for quickscore applied to a probabilistic version of Quick Medical Reference (QMR) are provided.

We propose a norm of consistency for a mixed set of defeasible and strict sentences, based on a probabilistic semantics. This norm establishes a clear distinction between knowledge bases depicting exceptions and those containing outright contradictions. We then define a notion of entailment based also on probabilistic considerations and provide a characterization of the relation between consistency and entailment. We derive necessary and sufficient conditions for consistency, and provide a simple decision procedure for testing consistency and deciding whether a sentence is entailed by a database. Finally, it is shown that if al1 sentences are Horn clauses, consistency and entailment can be tested in polynomial time.

This paper presents an approach for developing the explanation capabilities of rule-based expert systems managing imprecise and uncertain knowledge. The treatment of uncertainty takes place in the framework of possibility theory where the available information concerning the value of a logical or numerical variable is represented by a possibility distribution which restricts its more or less possible values. We first discuss different kinds of queries asking for explanations before focusing on the two following types : i) how, a particular possibility distribution is obtained (emphasizing the main reasons only) ; ii) why in a computed possibility distribution, a particular value has received a possibility degree which is so high, so low or so contrary to the expectation. The approach is based on the exploitation of equations in max-min algebra. This formalism includes the limit case of certain and precise information.

Bayesian inference systems should be able to explain their reasoning to users, translating from numerical to natural language. Previous empirical work has investigated the correspondence between absolute probabilities and linguistic phrases. This study extends that work to the correspondence between changes in probabilities (updates) and relative probability phrases, such as "much more likely" or "a little less likely." Subjects selected such phrases to best describe numerical probability updates. We examined three hypotheses about the correspondence, and found the most descriptively accurate of these three to be that each such phrase corresponds to a fixed difference in probability (rather than fixed ratio of probabilities or of odds). The empirically derived phrase selection function uses eight phrases and achieved a 72% accuracy in correspondence with the subjects' actual usage.

As the technology for building knowledge based systems has matured, important lessons have been learned about the relationship between the architecture of a system and the nature of the problems it is intended to solve. We are implementing a knowledge engineering tool called BART that is designed with these lessons in mind. BART is a Bayesian reasoning tool that makes belief networks and other probabilistic techniques available to knowledge engineers building classificatory problem solvers. BART has already been used to develop a decision aid for classifying ship images, and it is currently being used to manage uncertainty in systems concerned with analyzing intelligence reports. This paper discusses how state-of-the-art probabilistic methods fit naturally into a knowledge based approach to classificatory problem solving, and describes the current capabilities of BART.

Two major difficulties in using default logics are their intractability and the problem of selecting among multiple extensions. We propose an approach to these problems based on integrating nommonotonic reasoning with plausible reasoning based on triangular norms. A previously proposed system for reasoning with uncertainty (RUM) performs uncertain monotonic inferences on an acyclic graph. We have extended RUM to allow nommonotonic inferences and cycles within nonmonotonic rules. By restricting the size and complexity of the nommonotonic cycles we can still perform efficient inferences. Uncertainty measures provide a basis for deciding among multiple defaults. Different algorithms and heuristics for finding the optimal defaults are discussed.