A Model of High-Dimensional Feature Reduction Based on Variable Precision Rough Set and Genetic Algorithm in Medical Image

Table 4

The statistical results of the first group for experiment 1 with the SVM classifier.

Experiment times		Accuracy (%)	Sensitivity (%)	Specificity (%)	Time (s)
Reduction 1	1	92.50	86.00	99.00	33.2607
	2	97.25	97.50	97.00	34.5051
	3	99.00	99.00	99.00	35.8677
	4	98.00	99.50	96.50	35.1075
	5	99.25	100.00	98.50	35.5948
	Average value	97.20	96.40	98.00	34.8672
Reduction 2	1	93.25	87.00	99.50	43.6533
	2	97.25	98.00	96.50	44.5900
	3	98.75	98.50	99.00	43.9506
	4	97.50	99.00	96.00	44.5064
	5	99.25	100.00	98.50	44.8513
	Average value	97.20	96.50	97.90	44.3103
Reduction 3	1	92.75	87.50	98.00	32.6916
	2	95.75	97.50	94.00	33.7209
	3	98.00	98.00	98.00	34.0701
	4	97.00	98.50	95.50	33.3174
	5	98.50	99.50	97.50	33.3257
	Average value	96.40	96.20	96.60	33.4251
Reduction 4	1	93.75	88.50	99.00	30.6485
	2	96.50	97.50	95.50	31.9240
	3	97.50	97.00	98.00	32.4082
	4	97.75	99.00	96.50	32.5419
	5	99.00	100.00	98.00	36.1516
	Average value	96.90	96.40	97.40	31.8806
Reduction 5	1	94.00	89.00	99.00	31.9557
	2	96.50	97.50	95.50	33.6909
	3	97.25	97.00	97.50	33.0159
	4	97.75	99.00	96.50	39.9980
	5	99.00	100.00	98.00	32.5692
	Average value	96.90	96.50	97.30	34.2459