THEORETICAL FRAMEWORKS FOR PATTERN RECOGNITION PROBLEMS.
Abstract
This report summarizes several studies conducted on pattern recognition problems. One of these involved two problems of parametric statistics with a view to their application to nonsupervised pattern recognition. Another examined the question of the least costly word equivalent to a given word in the case in semigroups in which the equivalence relation is a two-sided congruence relation. A third continues a study of current flow in the central nervous system. The quantitative relationship between the parameter of stimulus current intensity and the anatomical characteristics of certain neural elements in the nervous system was examined. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 1967
- Accession Number
- AD0666473
Entities
People
- Wilson P. Tanner Jr
Organizations
- University of Michigan