| Abstract
|
A domain-independent knowledge-based infer … A domain-independent knowledge-based inference structure is presented, specificto the task of abstracting higher-level concepts from time-stamped data. Theframework includes a model of time, parameters, events, and contexts. A formalspecification of a domain's temporal-abstraction knowledge supportsacquisition, maintenance, reuse, and sharing of that knowledge.The knowledge-based temporal-abstraction method decomposes thetemporal-abstraction task into five subtasks. These subtasks are solved byfive domain-independent temporal-abstraction mechanisms. Thetemporal-abstraction mechanisms depend on four domain-specific knowledge types:structural, classification (functional), temporal-semantic (logical), andtemporal-dynamic (probabilistic) knowledge. Values for the four knowledgetypes are specified when developing a temporal-abstraction system in a newdomain. The knowledge-based temporal-abstraction method has been implemented in theRESUME system and evaluated in several different clinical domains(protocol-based care, monitoring of children's growth, and therapy of diabetes)with encouraging results. rapy of diabetes)with encouraging results.
|
