Accession Number:

ADA172866

Title:

A Framework for Empirical Discovery.

Descriptive Note:

Interim rept. May-Aug 86,

Corporate Author:

CALIFORNIA UNIV IRVINE DEPT OF INFORMATION AND COMPUTER SCIENCE

Personal Author(s):

• Langley,Pat
• Nordhausen,Bernd

Report Date:

1986-09-24

Pagination or Media Count:

35.0

Abstract:

Previous research in machine learning has viewed the process of empirical discovery as search through a space of theoretical terms. This paper proposes a problem space for empirical discovery, specifying six complementary operators for defining new terms that ease the statement of empirical laws. The six types of terms include numeric attributes such as PVT intrinsic properties such as mass composite objects such as pairs of colliding balls classes of objects such as acids and alkalis composite relations such as chemical reactions and classes of relations such as combustionoxidation. We review existing machine discovery systems in light of this framework, examining which parts of the problem space were covered by these systems. Finally, we outline an integrated discovery system IDS we are constructing that includes all six of the operators and which should be able to discover a broad range of empirical laws.

Descriptors:

• *ARTIFICIAL INTELLIGENCE
• MASS
• DISPLACEMENT
• THEORY
• CLUSTERING
• CHEMICAL REACTIONS
• HEURISTIC METHODS
• SALTS
• TAXONOMY
• FLUID FLOW
• ACIDS
• IDEAL GAS LAW

Subject Categories:

Distribution Statement:

APPROVED FOR PUBLIC RELEASE