Table 3. Results of the proposed methods compared to other methods from the literature on the Banana dataset.

Method		Reduced (ratio)	Accuracy	Training Speedup	Testing Speedup	Reduced SVs (ratio)
Whole dataset		1.00	0.896	1.00	1.00	1.00
FIFDR [13]		0.684	0.877	4.99	1.741	0.315
CBCH (K = 100) [15]		0.70625000	0.899057	1.540797	0.968350	1.024043
BPLSH (M = 90, L = 10) [16]		0.24707547	0.881887	5.138549	1.077517	0.857766
Proposed Methods	DBI (r = 0.5)	0.5	0.900	1.98	0.96	1.028
	BRI (r = 0.2)	0.2	0.893	9.36	2.12	0.47
	BRIX (r = 0.1)	0.1	0.88	51.62	11.7	0.068
	SVO (k = 1)	0.2217	0.898	5.23	0.97	0.996
	SVOX (k = 1)	0.1335	0.902	17.14	3.28	0.26

The first row represents the metrics measured after training on the whole dataset. “Reduced (ratio)” is the ratio of the size of the resulting reduced dataset to the size of the whole dataset. “Reduced SVs (ratio)” is the ratio of the number of SVs of the reduced dataset to that of the whole dataset.