An Efficient Cost-Sensitive Feature Selection Using Chaos Genetic Algorithm for Class Imbalance Problem

Table 6

Classification evaluation results for the KNN and C4.5 algorithm with feature selection (NON, CFS, CASH).

Algorithm		F-measure
Feature selection	Classifier	Normal	DOS	PROBE	R2L	U2R
—	KNN	0.996	0.998	0.990	0.936	0.876
CFS	KNN	0.997	0.998	0.990	0.938	0.899
CASH	KNN	0.999	0.998	0.990	0.936	0.866
CSFSG	KNN	0.997	0.997	0.989	0.942	0.916
—	C4.5	0.997	0.998	0.988	0.963	0.966
CFS	C4.5	0.997	0.998	0.988	0.967	0.966
CASH	C4.5	0.997	0.989	0.969	0.942	0.963
CSFSG	C4.5	0.997	0.998	0.953	0.968	0.966
Algorithm		Recall
Feature selection	Classifier	Normal	DOS	PROBE	R2L	U2R
—	KNN	0.996	0.998	0.986	0.779	0.6
CFS	KNN	0.997	0.998	0.78	0.724	0.8
CASH	KNN	0.999	0.998	0.971	0.75	0.4
CSFSG	KNN	0.997	0.998	0.964	0.794	0.6
—	C4.5	0.997	0.998	0.971	0.735	0.4
CFS	C4.5	0.997	0.998	0.788	0.75	0.2
CASH	C4.5	0.997	0.989	0.967	0.75	0.4
CSFSG	C4.5	0.997	0.998	0.969	0.738	0.6
Algorithm		ROC area
Feature selection	Classifier	Normal	DOS	PROBE	R2L	U2R
—	KNN	0.996	0.998	0.990	0.983	0.899
CFS	KNN	0.999	0.998	0.996	0.977	0.99
CASH	KNN	0.999	0.998	0.988	0.938	0.899
CSFSG	KNN	0.998	0.999	0.993	0.947	0.988
—	C4.5	0.999	0.998	0.992	0.937	0.579
CFS	C4.5	0.997	0.998	0.963	0.886	0.733
CASH	C4.5	0.997	0.989	0.998	0.887	0.721
CSFSG	C4.5	0.997	0.998	0.996	0.935	0.733