Representing and reasoning about physical systems using prime models
From Semantic Portal Wiki
{{#vardefine:category|Publication}}{{#vardefine:templatename|i.publication}}{{#vardefine:package|smwbp_instance_templates}}
| Edit |
Reference: {{#vardefine:pagename|representing and reasoning about physical systems using prime models }}
- [[]]
bibtex
{{#vardefine:pagename|Representing and reasoning about physical systems using prime models }}{{#vardefine:key| }}
abstract: We propose an approach based on a network formalism for explicitlyrepresenting knowledge about physical systems at two levels of abstraction.Prime models explicitly represent the abstract structures and processes, bothnormal and abnormal, underlying classes pf physical systems. Domain modelsexplicitly represent the actual structures and processes that make upparticular systems. Each domain model is viewed as an instance of aparticular prime model. This approach has several advantages. It provides abasis for reasoning from first principles about individual domain models andyields building blocks for reasoning about more complex systems. It offers acompact representation of a potentially very large body of knowledge availablefor use in various reasoning tasks. In real world applications we often haveto deal with uncertain and incomplete information or domains whereprobabilistic reasoning is more appropriate. Thus, we explore a beliefnetwork, a well known network used for representing and reasoning based onprobabilistic theories. We discuss the tradeoff between the proposed approachand the belief network and show how we can use prime models to represent andreason about physical systems under uncertainty.
download:
- paper:
- slides:
| Abstract | We propose an approach based on a network … We propose an approach based on a network formalism for explicitlyrepresenting knowledge about physical systems at two levels of abstraction.Prime models explicitly represent the abstract structures and processes, bothnormal and abnormal, underlying classes pf physical systems. Domain modelsexplicitly represent the actual structures and processes that make upparticular systems. Each domain model is viewed as an instance of aparticular prime model. This approach has several advantages. It provides abasis for reasoning from first principles about individual domain models andyields building blocks for reasoning about more complex systems. It offers acompact representation of a potentially very large body of knowledge availablefor use in various reasoning tasks. In real world applications we often haveto deal with uncertain and incomplete information or domains whereprobabilistic reasoning is more appropriate. Thus, we explore a beliefnetwork, a well known network used for representing and reasoning based onprobabilistic theories. We discuss the tradeoff between the proposed approachand the belief network and show how we can use prime models to represent andreason about physical systems under uncertainty. about physical systems under uncertainty. |
| Author | Rattikorn Hewett +, and Barbara Hayes-Roth + |
| Bibtype | article + |
| Journal | Principles of Semantic Networks + |
| Key | KSL-90-40 + |
| Publisher | Morgan Kaufmann + |
| Tag | Computer science + |
| Title | Representing and reasoning about physical systems using prime models + |
| Tr id | KSL-90-40 + |
| Year | 1990 + |
