A Discrete Crow Search Algorithm for Mining Quantitative Association Rules
No Thumbnail Available
Date
2021
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Associationrulesarethespecificdataminingmethodsaimingtodiscoverexplicitrelationsbetweenthe
differentattributesinalargedataset.However,inreality,severaldatasetsmaycontainbothnumericand
categoricalattributes.Recently,manymeta-heuristicalgorithmsthatmimicthenaturearedeveloped
forsolvingcontinuousproblems.Thisarticleproposesanewalgorithm,DCSA-QAR,formining
quantitativeassociationrulesbasedoncrowsearchalgorithm(CSA).Toaccomplishthis,newoperators
aredefinedtoincreasetheabilitytoexplorethesearchingspaceandensurethetransitionfromthe
continuoustothediscreteversionofCSA.Moreover,anewdiscretizationalgorithmisadoptedfor
numericalattributestakingintoaccountdependenciesprobablythatexistbetweenattributes.Finally,
toevaluatetheperformance,DCSA-QARiscomparedwithparticleswarmoptimizationandmono
andmulti-objectiveevolutionaryapproachesforminingassociationrules.Theresultsobtainedover
real-worlddatasetsshowtheoutstandingperformanceofDCSA-QARintermsofqualitymeasures.