Project Title:

A Knowledge-Based System for Analyzing Technical Data

92-1-06.01-9457                                                   NAS01-19893
A Knowledge-Based System for Analyzing Technical
     Data
Nielsen Engineering & Research, Inc.
510 Clyde Avenue
Mountain View, CA  94043-2287
Laura C. Rodman                                      (415-968-9457)

Abstract:

     Recent technical advances have made numerical simulations of
physical phenomena very efficient at generating large amounts of
data. In some fields, the inability to analyze all the collected
data may impede the understanding of increasingly complex
phenomena. The goal of this project is to produce an automated
technique that will filter large amounts of data in order to
identify interesting trends and relationships. This approach uses
knowledge-based systems to replicate the reasoning steps that
would be taken by an investigator to find trends and relationships
in a set of data. For example, the program can look at the
sensitivity of a quantity to other variables in the data, or for
strong correlations among specified quantities. The automation of
this process would allow very large amounts of data to be studied
much more rapidly than is currently possible, and may allow the
discovery of certain trends that might have been otherwise missed.
This system will filter through data more quickly in order to
pinpoint areas of interest so that the scientist can concentrate
on studying those particular areas.

Potential Commercial Application:

Potential Commercial Applications:  This project addresses a
critical need to analyze more efficiently large quantities of
data. Potential aerospace applications are in computational fluid
dynamics, satellite imaging, and ground testing.

                                     ***