Research of Financial Early-Warning Model on Evolutionary Support Vector Machines Based on Genetic Algorithms

Table 2

Classification accuracies of various parameters in the first model.

	𝛿^2
	1		10		30		50		80
C	train	Test	train	test	train	test	train	test	train	test
1	86.87	87.50	80.00	70.83	66.67	70.83	66.67	70.83	66.67	70.83
10	93.33	75.50	86.67	70.83	66.67	70.83	66.67	70.83	66.67	70.83
30	93.33	75.50	86.67	70.83	66.67	70.83	66.67	70.83	66.67	75.00
50	100	79.17	86.67	79.17	66.67	70.83	66.67	70.83	66.67	75.00
90	100	79.17	93.33	87.50	66.67	70.83	66.67	70.83	66.67	75.00
100	100	79.17	93.33	79.17	66.67	70.83	66.67	79.17	66.67	70.83
150	100	79.17	86.67	79.17	66.67	70.83	66.67	79.17	66.67	70.83
200	100	79.17	86.67	75.00	66.67	70.83	66.67	70.83	66.67	70.83
250	100	79.17	86.67	79.17	66.67	70.83	66.67	70.83	66.67	70.83