×

A computational approach to approximate and plausible reasoning with applications to expert systems. (English) Zbl 0565.68089

A survey paper on various kinds of approximate and plausible reasoning with a large list of references.
Uncertain, imprecise, incomplete or even inconsistent knowledge is now widely discussed especially in the area of knowledge engineering. The author surveys several mathematical models of uncertainty as Bayesian model, Shafer’s belief theory, Zadeh’s possibility and fuzzy set theories, etc. A common basis for these approaches is presented with its consequences to inexact reasoning, i.e. to deductive inferencing with both weighted and imprecise (or fuzzy) premises. Such techniques play an important role in various well-known knowledge based systems like MYCIN, PROSPECTOR, CASNET and others. Thus the paper should be of interest to anyone concerned with automated reasoning techniques in expert systems.
The problem of uncertainty and of inexact reasoning is not, of course, yet definitely settled. Various alternative, especially nonprobabilistic approaches are intensively investigated.
Reviewer: P.Jirků

MSC:

68T99 Artificial intelligence
03B48 Probability and inductive logic
03B52 Fuzzy logic; logic of vagueness
68T15 Theorem proving (deduction, resolution, etc.) (MSC2010)
03B50 Many-valued logic
68Q65 Abstract data types; algebraic specification
68-02 Research exposition (monographs, survey articles) pertaining to computer science
PDFBibTeX XMLCite
Full Text: DOI