Tree structures of linear threshold units for the classification of numeric examples

Miroslav Kubat, Doris Flotzinger

Research output: Contribution to journalArticle

1 Scopus citations

Abstract

A decision tree-based system for learning from numeric data is described. Results from linear algebra (pseudoinverse matrices) help the system to generate decision trees where the nodes are represented by linear threshold units minimizing the mean square error. The system's capability to provide good classifications with small decision trees is demonstrated on artificial and benchmark data.

Original languageEnglish (US)
Pages (from-to)521-533
Number of pages13
JournalCybernetics and Systems
Volume26
Issue number5
DOIs
StatePublished - Jan 1 1995
Externally publishedYes

    Fingerprint

ASJC Scopus subject areas

  • Software
  • Information Systems
  • Artificial Intelligence

Cite this