Adapting to drift in continuous domains (Extended abstract)

Miroslav Kubat, Gerhard Widmer

Research output: Chapter in Book/Report/Conference proceedingConference contribution

36 Scopus citations

Abstract

The experiments demonstrate that FRANN compares favourably with FLORA4 in the presence of concept drift. Learning is possible from examples described by symbolic as well as by numeric attributes, and because of its representation formalism (RBF networks, which realize a kind of prototype weighting scheme) FRANN is particularly effective in capturing concepts with nonlinear boundaries.

Original languageEnglish (US)
Title of host publicationMachine Learning
Subtitle of host publicationECML-95 - 8th European Conference on Machine Learning, 1995, Proceedings
EditorsNada Lavrac, Stefan Wrobel
PublisherSpringer Verlag
Pages307-310
Number of pages4
ISBN (Print)3540592865, 9783540592860
DOIs
StatePublished - 1995
Event8th European Conference on Machine Learning, ECML 1995 - Heraclion, Greece
Duration: Apr 25 1995Apr 27 1995

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume912
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other8th European Conference on Machine Learning, ECML 1995
CountryGreece
CityHeraclion
Period4/25/954/27/95

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Adapting to drift in continuous domains (Extended abstract)'. Together they form a unique fingerprint.

Cite this