Adapting to drift in continuous domains (Extended abstract)

Miroslav Kubat, Gerhard Widmer

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

36 Scopus citations


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
Number of pages4
ISBN (Print)3540592865, 9783540592860
StatePublished - 1995
Externally publishedYes
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)
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Other8th European Conference on Machine Learning, ECML 1995

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)


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

Cite this