Table 2. Results of the proposed SVO & SVOX methods on the Banana dataset with different values of k.

k		Reduced (ratio)		Training Speedup		Testing Speedup		Accuracy		Number of SVs		Jaccard similarity
Whole dataset		1.00		1.00		1.00		0.896		940		1.00
	SVO	SVOX	SVO	SVOX	SVO	SVOX	SVO	SVOX	SVO	SVOX	SVO	SVOX
1	0.22	0.13	5.23	17.14	0.97	3.28	0.898	0.902	937	244	1.00	0.26
2	0.24	0.15	4.00	15.53	0.96	2.98	0.895	0.901	939	245	0.98	0.26
3	0.26	0.17	3.44	12.85	0.97	2.45	0.895	0.901	938	333	0.98	0.35
4	0.27	0.19	4.08	10.78	0.96	1.95	0.901	0.901	973	334	0.95	0.35
5	0.29	0.20	3.79	11.67	0.94	2.42	0.901	0.901	973	334	0.95	0.35
10	0.35	0.27	2.96	8.66	0.96	2.44	0.901	0.899	975	340	0.94	0.35
15	0.40	0.32	2.62	7.86	0.96	2.38	0.901	0.900	977	344	0.94	0.36
20	0.44	0.36	2.41	6.92	0.96	2.40	0.900	0.901	977	346	0.94	0.36
30	0.51	0.43	2.11	5.87	0.98	2.44	0.901	0.898	977	347	0.94	0.36
40	0.57	0.50	1.80	5.04	0.93	2.33	0.901	0.898	980	348	0.94	0.36
50	0.62	0.55	1.73	4.67	0.92	2.49	0.901	0.898	982	345	0.94	0.36

Each row compares performance metrics of the methods for a different value (1 to 50) of the parameter k of the SVO & SVOX methods. 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.