ERIC Number: ED310166
Record Type: Non-Journal
Publication Date: 1988-Dec-27
Reference Count: N/A
Learning from Physical Analogies: A Study in Analogy and the Explanation Process.
Falkenhainer, Brian Carl
Analogical reasoning and learning applied to the task of constructing qualitative explanations for physical phenomena are the subjects of this investigation. Two issues are addressed. The first is how analogies are elaborated to sanction new inferences about a novel situation. This issue is adddressed by contextual structure-mapping, a knowledge-intensive adaptation of D. Gentner's structure-mapping theory. This approach presents analogy elaboration as a map-and-analyze cycle, in which two situations are placed in correspondence, followed by problem solving and inference production focused on correspondence inadequacies. The second issue is the evaluation the quality of a proposed analogy and its use for some performance task. A theory of verification-based analogical learning is presented to address the tenuous nature of analogically inferred concepts, and procedures for increasing confidence in the inferred knowledge are described. Specifically, it relies on analogical inference to hypothesize new theories and simulation of those theories to analyze their validity. It represents a view of analogy as an iterative process of hypothesis formation, testing, and revision. These ideas are illustrated via PHINEAS, a computer program that uses similarity to posit qualitative explanations for time-varying descriptions of physical behaviors. It builds upon existing work in qualitative physics to provide a means with which to describe and reason with theories of the physical world. A 144-item list of references is included. (TJH)
Publication Type: Reports - Evaluative; Dissertations/Theses - Doctoral Dissertations
Education Level: N/A
Sponsor: Office of Naval Research, Arlington, VA. Personnel and Training Research Programs Office.
Authoring Institution: Illinois Univ., Urbana. Dept. of Computer Science.