Automated model selection using context-dependent behaviors

From Semantic Portal Wiki

Jump to: navigation, search

{{#vardefine:category|Publication}}{{#vardefine:templatename|i.publication}}{{#vardefine:package|smwbp_instance_templates}}

Edit

Reference: {{#vardefine:pagename|automated model selection using context-dependent behaviors }}

  1. [[]]

bibtex

{{#vardefine:pagename|Automated model selection using context-dependent behaviors }}{{#vardefine:key| }}

abstract: Effective problem-solving about complex engineered devices requires devicemodels that are both adequate for the problem and computationally efficient.Producing such models requires identifying the relevant device features anddetermining applicable simplifications. This paper presents a method forautomatically constructing a device model by selecting an appropriate modelfor each of the device’s components using the context in which it operates.We introduce context-dependent behaviors (CDBs), a frame-like componentbehavior model representation for encapsulating contextual modelingconstraints. We show how CDBs are used in the model selection process byexploiting constraints from three sources: the structural and behavioralcontexts of the components, and the expected behavior of the device. Wedescribe an implemented program for model selection. The inputs are thestructure of the device--the components of the device and structural relationsbetween them--the expected device behavior, and a library of CDBs. The outputis a set of component CDBs forming a structurally and behaviorally consistentdevice model that achieves the expected behavior. We demonstrate the programon a temperature gauge.

download:

  • paper:
  • slides:
Facts about Automated model selection using context-dependent behaviorsRDF feed
AbstractEffective problem-solving about complex en Effective problem-solving about complex engineered devices requires devicemodels that are both adequate for the problem and computationally efficient.Producing such models requires identifying the relevant device features anddetermining applicable simplifications. This paper presents a method forautomatically constructing a device model by selecting an appropriate modelfor each of the device’s components using the context in which it operates.We introduce context-dependent behaviors (CDBs), a frame-like componentbehavior model representation for encapsulating contextual modelingconstraints. We show how CDBs are used in the model selection process byexploiting constraints from three sources: the structural and behavioralcontexts of the components, and the expected behavior of the device. Wedescribe an implemented program for model selection. The inputs are thestructure of the device--the components of the device and structural relationsbetween them--the expected device behavior, and a library of CDBs. The outputis a set of component CDBs forming a structurally and behaviorally consistentdevice model that achieves the expected behavior. We demonstrate the programon a temperature gauge. nstrate the programon a temperature gauge.
AuthorP. P  +, Urang Nayak  +, Leo Joskowicz  +, and Sanjaya Addanki  +
Bibtypetechreport  +
InstitutionKnowledge Systems, AI Laboratory  +
KeyKSL-91-16  +
MonthMay  +
NumberKSL-91-16  +
TagComputer science  +
TitleAutomated Model Selection using Context-Dependent Behaviors  +
Tr idKSL-91-16  +
Year1991  +
Personal tools
Semantic Web Community
Tetherless World constellation
maintenance