Selection of Variables in Dlustering and Pattern Recognition.
Abstract
The report discusses the following: The method of SELFIC which extracts major discriminating features from a collection of object; The method of CLAFIC which assigns subspaces instead of volumes to classes, thus introducing a new method of discrimination; The universal clustering to bilateral or multilateral relations; The well-balanced decision function which excels the other known decision functions also to the speed of convergence, and the method of object-predicate reciprocation. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 01, 1972
- Accession Number
- AD0737763
Entities
People
- Satosi Watanabe
Organizations
- University of Hawaiʻi System