Research on Parallel Support Vector Machine Based on Spark Big Data Platform

Table 3

Training time.

Number	Dataset	LIBSVM (s)	Cascade SVM (s)	Ours (s)
1	Dataset1	31.30	46.16	44.57
2	Dataset2	42.03	49.41	47.03
3	Dataset3	94.23	63.47	59.06
4	Dataset4	25.10	18.21	16.42
5	Dataset5	1042.73	541.43	446.83
6	Dataset6	143.44	30.25	24.73
7	Dataset7	30144.50	2503.07	1601.30