Supervised methods of machine learning for email classification: a literature survey
Nov 20, 2025
DOI:
Published in: Systems Science & Control Engineering
Publisher: Taylor & Francis
Intoday’sdigital landscape,email isacknowledgedasacriticalconduitforglobaldataexchanges. Withasurgeindatavolume,malefactorsexploituseridentities,leadingtodatamisuse.Cybercriminalsemployelectronictransgressionssuchasphishingandspamtoorchestratesecurityinfractions. Machinelearningcountersthesebreachesusingmyriadtechniques,demonstratingsignificantefficiencyinidentifyingphishingemails.Wecandividemachinelearningintotwotypes:supervised andunsupervised.Supervisedlearningrequirespre-trainingthemodelonlabelleddatasets,amalgamatingclassification,andregressionlearning.Notably,supervisedmethodologiessuchassupport vectormachines (SVMs),naiveBayes,decisiontrees,neuralnetworks, randomforests,anddeep learninghavebeenexploitedforspamfiltering.Thisreviewdelvesintoissuesconcerningspamfilteringandemailclassificationthroughsupervisedmachinelearningtechniques,offeringacomprehensiveevaluationofstrategies,methods,performanceindicators,andthebenefitsanddrawbacks ofdifferent research. This informationallows researchers toassess theefficiencyandeffectivenessofsupervisedlearningalgorithms, layingthefoundationforadvancedemailcategorization techniques.