Empirical study of seven data mining algorithms on different characteristics of datasets for biomedical classification applications

Table 6 Summary of applicative algorithm recommendation on different characteristic datasets

Character of dataset	NB	LR	kNN	C4.5	SVM	AB	RF	Represents of dataset
Small sample size	√	√				√	√	Iris, wine
High correlation	√	√						Iris, wine
Binary-class task		√			√	√		Breast cancer Wisconsin, Wdbc
Balanced data	√	√			√			Wine, breast cancer Wisconsin, Wdbc
Multi-class task			√	√			√	Abalone, wine quality_red
Imbalanced data			√	√			√	Wine quality_white
Large sample size			√				√	Adult, poker hand
Low correlation			√	√		√	√	Car evaluation, Wpbc, heart disease

ISSN: 1475-925X