Determining Arguments of Invariant Functional Descriptions

  • Authors:
  • Mieczyslaw M. Kokar

  • Affiliations:
  • Department of Industrial Engineering and Information Systems, Northeastern University, Boston, MA 02115, U.S.A. KOKAR@NORTHEASTERN.CSNET

  • Venue:
  • Machine Learning
  • Year:
  • 1986

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper we examine the problem of determining arguments of invariant functional descriptions from incomplete observational data. Physical laws are one example of invariant functional descriptions. For such functions, we show that one can test the relevance of the function's arguments even though their values remain constant throughout the observational data. We present a method, called COPER, for discovering invariant functional descriptions. COPER eliminates irrelevant arguments, generates additional relevant arguments, and generates a functional formula. We focus on the first two of these features, giving two examples of how the methodology can be applied to determining arguments of physical laws.