Models for reasoning with multitype uncertainty in expert systems
Uncertainty models play an important role within expert systems. However, there are different types of uncertainty (inaccuracy, inexactitude, fuzziness etc.). It can be shown that the various uncertainty models as known from literature in fact are dealing with different types of uncertainty. The type of uncertainty, which is characteristic for the application for which the expert system is used, has a direct impact on the selection of the appropriate uncertainty model within a given application domain.
Problems appear when within an application domain various types of uncertainty should be handled at the same time (multitype uncertainty). In this case a special inference calculus for e.g. the combination of evidences, related to various types of uncertainty, is needed. In this paper two general methods for an inference calculus for multitype uncertainty will be proposed and evaluated.
KeywordsExpert systems multitype uncertainty uncertainty classes and types uncertainty calculi inference calculi certainty vector rule inference classification inference rule generation
Unable to display preview. Download preview PDF.
- Backer E., Van der Lubbe J.C.A., Krijgsman W. (1988), On modelling of uncertainty and inexactness in expert systems, Proc. Ninth Symp. on Information Theory, Mierlo, the Netherlands, pp. 105–111Google Scholar
- Backer E., Gerbrands J.J., Bloom G., Reiber J.H.C., Reijs A.E.M., Van den Herik H.J. (1988), Developments towards an expert system for the quantitive analysis of thallium-201 scintigrams, In: De Graaf, C.N., Viergever, M.A., Eds., Information Processing in Medical Imaging, New York, pp. 293–306Google Scholar
- Bonissone P.P., Tong R.M. (1985), Reasoning with uncertainty in expert systems, Int. J. Man-Machine Studies, 22, pp. 241–250Google Scholar
- Buchanan B.G., Shortliffe E.H. (1984), Rule-based expert systems, Massachusetts, 1984Google Scholar
- Dubois D., Prade H. (1987), A tentative comparison of numerical approximate reasoning methodologies, Int. J. Man-Machine Studies, 27, pp. 717–728Google Scholar
- Prade H. (1985), A computational approach to approximate and plausible reasoning with applications to expert systems, IEEE Pattern Anal. Mach. Intell., Vol PAMI-7, 3, pp. 284–298Google Scholar
- Shafer G. (1975), A mathematical theory of evidence, Princeton Univ. PressGoogle Scholar