251. CERTAINTY FACTOR THEORY AND ITS IMPLEMENTATION IN AN INFORMATION SYSTEM-ORIENTED EXPERT SYSTEM SHELL
- Author
-
Dan Qiu
- Subjects
business.industry ,Computer science ,Mycin ,media_common.quotation_subject ,Probabilistic logic ,Context (language use) ,Certainty ,computer.software_genre ,Expert system ,Interpretation (model theory) ,Knowledge base ,Artificial Intelligence ,Argument ,Artificial intelligence ,business ,computer ,Mathematical economics ,media_common - Abstract
This article discusses some topics related to the MYCIN model for uncertainty handling in expert systems. Some previously identified problems in this model should be considered as evidence to support the point of view that the probabilistic interpretation for measure of belief (MB) and measure of disbelief (MD) given by Shortliffe is inappropriate. In connection lo this, some results achieved by Heckerman are also reviewed and one of his arguments is questioned. According to this argument, the original interpretation for certainty factors implies a noncommutative method for evidence combination, and therefore, this interpretation should be abandoned. Problems similar to those identified by Adams in the original interpretation for certainty factors exist also in the interpretation recommended by Heckerman. Then, in the context of previous results stating that there is an isomorphic mapping from the evidence combination scheme for the likelihood ratio to the evidence combination function adopted in EMYCIN a...
- Published
- 1992
- Full Text
- View/download PDF