Journal
SWARM AND EVOLUTIONARY COMPUTATION
Volume 69, Issue -, Pages -Publisher
ELSEVIER
DOI: 10.1016/j.swevo.2021.101006
Keywords
Machine learning; Single-solution-based metaheuristics; Evolutionary algorithms; Swarm intelligence methods
Ask authors/readers for more resources
This paper presents a state-of-the-art review of using single-solution-based metaheuristics and swarm and evolutionary computational algorithms to build decision trees as classification models, outlining the decision-tree induction process components and detailing existing literature studies on metaheuristic-based approaches to building these classifiers. A summary analysis of these studies is also conducted, focusing on their internal components and experimental studies.
The induction of decision trees is a widely-usedapproachtobuildclassificationmodelsthatguaranteehighperfor-manceandexpressiveness.Sincearecursive-partitioningstrategyguidedforsomesplittingcriterioniscommonlyusedtoinducetheseclassifiers,overfitting,attributeselectionbias,andinstabilitytosmalltrainingsetchangesarewell-knownproblemsinthem.Otherapproaches,suchasincrementalinduction,classifierensembles,andtheglobalsearchinthedecision-tree-space,havebeenimplementedtoovercometheseproblems.Inparticular,metaheuristicssuchassimulatedannealing,geneticalgorithms,geneticprogramming,andantcolonyoptimiza-tionhavebeenusedtoinducecompactandaccuratedecisiontrees.Thispaperpresentsastate-of-the-artreviewoftheuseofsingle-solution-basedmetaheuristicsandswarmandevolutionarycomputationalgorithmstobuilddecisiontreesasclassificationmodels.Weoutlinethedecision-tree-inductionprocesscomponentsanddetailtheexistingliteraturestudiesonmetaheuristic-basedapproachestobuildingtheseclassifiers.Severaltimelinesshow-ingthechronologicalorderinwhichtheseapproacheswereintroducedintheliteratureareincluded.Asummaryanalysisofthesestudiesisalsoconducted,focusingontheirinternalcomponentsandexperimentalstudies.Thisworkprovidesausefulreferencepoint for future research in this field
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available