A Lightweight Data Preprocessing Strategy with Fast Contradiction Analysis for Incremental Classifier Learning

Table 1

Performance comparison of algorithms under batch learning mode.

Algorithm	Batch learning
	No preprocessing			Preprocessing
	Accuracy	Time	Gain ratio	Accuracy	Time	Gain ratio
J48	62.5110	33.0000	1.8943	99.7983	9.0000	11.0887
NB	55.0207	3.0000	18.3402	88.0629	4.0000	22.0157
NN	60.0702	1083.0000	0.0555	93.2245	613.0000	0.1521
SVM	58.1060	3607.0000	0.0161	88.0324	803.0000	0.1096
Kstar	91.7483	7970.0000	0.0115	99.6041	7860.0000	0.0127
IBK	100.0000	258.0000	0.3876	100.0000	72.0000	1.3889
LWL	60.5204	16858.0000	0.0036	91.1804	5082.0000	0.0179
DT	61.5731	37.0000	1.6641	100.0000	42.0000	2.3810