Constructing the correct diagnosis when symptoms disappear
AAAI '98/IAAI '98 Proceedings of the fifteenth national/tenth conference on Artificial intelligence/Innovative applications of artificial intelligence
Information Sciences—Informatics and Computer Science: An International Journal
Minimal assumption distribution propagation in belief networks
UAI'93 Proceedings of the Ninth international conference on Uncertainty in artificial intelligence
IWINAC'05 Proceedings of the First international work-conference on the Interplay Between Natural and Artificial Computation conference on Artificial Intelligence and Knowledge Engineering Applications: a bioinspired approach - Volume Part II
Expert Systems with Applications: An International Journal
Hi-index | 0.00 |
A Bayesian method for the diagnostic classification of a singlecase into one of several classes based on qualitative data by an expertsystem is presented. Both the expert's knowledge about the base rates ofthe classes and about the conditional probabilities of the data withineach class are expressed by beta probability distributions. The Bayesianpoint probability that a single case belongs to a given class is treatedas a parameter. Its probability distribution is derived and called`credibility distribution'. Because the numerical handling of thedistribution is difficult, an approximation by a beta distribution isproposed. The credibility distribution expresses the imprecision of adiagnostic probability. It processes the amount of knowledge enteringthe probabilistic inference. It is shown that the total imprecision of adiagnostic classification can be decomposed into the sum of theimprecision of its knowledge components. Imprecision—notexpertise—is additive. In cascaded inference, the total degree ofdiagnostic expertise is also not additive, but recursively discounted ateach level of inference. The method allows one to work with subsets ofthe full problem space and thus reduces the combinatorialexplosion.—Author's Abstract