Skip to main navigation Skip to search Skip to main content

Bidirectional action rule learning

  • Silesian University of Technology
  • Institute of Innovative Technologies EMAG

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Citations (Scopus)

Abstract

Action rules specify recommendations which should be followed in order to transfer objects to the desired decision class. In this paper influence of employing information contained in source and target class examples in sequential covering based action rule induction method is examined. Results show that using source class for guiding the induction process produces best results.

Original languageEnglish
Title of host publicationComputer and Information Sciences - 32nd International Symposium, ISCIS 2018, Held at the 24th IFIP World Computer Congress, WCC 2018, Proceedings
EditorsErol Gelenbe, Ricardo Lent, Tadeusz Czachórski, Erol Gelenbe, Krzysztof Grochla
PublisherSpringer Verlag
Pages220-228
Number of pages9
ISBN (Print)9783030008390
DOIs
Publication statusPublished - 2018
Event32nd International Symposium on Computer and Information Sciences, ISCIS 2018 Held at the 24th IFIP World Computer Congress, WCC 2018 - Poznan, Poland
Duration: 20 Sept 201821 Sept 2018

Publication series

NameCommunications in Computer and Information Science
Volume935
ISSN (Print)1865-0929

Conference

Conference32nd International Symposium on Computer and Information Sciences, ISCIS 2018 Held at the 24th IFIP World Computer Congress, WCC 2018
Country/TerritoryPoland
CityPoznan
Period20/09/1821/09/18

Keywords

  • Action rules
  • Classification
  • Data mining

ASJC Scopus subject areas

  • General Computer Science
  • General Mathematics

Fingerprint

Dive into the research topics of 'Bidirectional action rule learning'. Together they form a unique fingerprint.

Cite this