Loading...
Please wait, while we are loading the content...
Similar Documents
Structures of rule-based belief functions (1986)
| Content Provider | CiteSeerX |
|---|---|
| Author | Eddy, William F. Pei, Gabriel P. |
| Abstract | Shafer’s theory of evidential reasoning has of the form “If A then B, ” where A and B are logical recently received much attention as a possible propositions.) There is currently a great deal of interest in model for probabilistic reasoning in expert system applications. This paper discusses the particular difficulties of implementing Shafer’s belief functions in the context of the most common form of expert system, rule-based systems. The two most important problems are: the representation of the expert’s subjective degrees of belief corresponding to his expressed rules, and the computational complexity of the inference mechanism for introducing uncertainty into the reasoning used in such production systems. Probability theory has been used as the basis for combining numerical measures of uncertainty in several rule-based expert systems, for example, the SRI system PROSPECTOR. PROSPECTOR assigned independent expert-supplied conditional probabilities to propositions (see [ 1 I). Users were permitted to input independent unconditional probabilities corresponding to observed combining evidence. We argue that a potential evidence. Bayes ’ rule was used to compute the posterior approach for dealing with both problems is given probabilities. by introducing constraints on the structure of the Certainty factors are an alternative scheme for modeling belief functions. These constraints, along with uncertainties in rule-based expert systems. (Consider the the expressed rules and the elicited belief MYCIN system for medical diagnosis described in [2].) values, form the expert’s total knowledge. Unlike probabilities, certainties are defined by and combined through an ad hoc set of rules. 1. |
| File Format | |
| Volume Number | 30 |
| Journal | IBM J. Res. Develop |
| Language | English |
| Publisher Date | 1986-01-01 |
| Access Restriction | Open |
| Subject Keyword | Rule-based Belief Function Independent Unconditional Probability Sri System Prospector Possible Proposition Expert System Application Rule-based Expert System Potential Evidence Shafer Belief Function Evidential Reasoning Probabilistic Reasoning Several Rule-based Expert System Important Problem Particular Difficulty Belief Function Alternative Scheme Expert Subjective Degree Much Attention Independent Expert-supplied Conditional Probability Medical Diagnosis Inference Mechanism Expert Total Knowledge Common Form Rule-based System Posterior Approach Certainty Factor Probability Theory Computational Complexity Expressed Rule Expert System Great Deal Elicited Belief Mycin System Ad Hoc Set Bayes Rule Production System Numerical Measure |
| Content Type | Text |
| Resource Type | Article |
