Accuracy and Efficiency in Pattern Classification.

Abstract

A pattern classification scheme which is grounded in classical probability theory may be associated with confidence internvals that represent an estimate of the predictive capability of the scheme. As a practical matter, realistic allocations of data acquisition and processing resources may severely constrain acceptable levels of predictability. We examine some of the basic assumptions which underlie the standard statistical techniques. In particular, we show that fuzzy logic effectively produces conservative estimates for the conditional probability of the union of sets since, in that case, it neglects information related to the intersection. We propose that such neglect can be remedied, at a computational cost, without resorting explicitly to the usual procedure of integrating over irregularly shaped volumes. To this end, we introduce a class of probability density distributions which possesses (hyper) rectangular contour. Explicit formulas for the normalization constant and the probability of error are then derived for typical distributions.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 1979
Accession Number
ADA066511

Entities

People

  • S. Berkowitz

Tags

Communities of Interest

  • Energy and Power Technologies
  • Human Systems
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Accuracy
  • Acquisition
  • Algorithms
  • Computations
  • Cost Reductions
  • Data Acquisition
  • Data Analysis
  • Eigenvalues
  • Fuzzy Logic
  • Fuzzy Sets
  • Information Science
  • Normal Distribution
  • Probability
  • Random Variables
  • Set Theory
  • Standards
  • Statistics

Fields of Study

  • Mathematics

Readers

  • Approximation Theory.
  • Artificial Intelligence
  • Computational Modeling and Simulation