Aero Engine Fault Diagnosis Using an Optimized Extreme Learning Machine

Table 2

Mean training and testing accuracy of six methods on different classification problems.

Algorithms	Satellite image		Image segmentation		Diabetes problem
	Training	Testing	Training	Testing	Training	Testing
Q-ELM	0.890	0.876	0.965	0.960	0.854	0.842
P-ELM	0.884	0.872	0.938	0.947	0.836	0.813
G-ELM	0.877	0.869	0.938	0.934	0.812	0.804
ELM	0.873	0.852	0.930	0.921	0.783	0.773
SVM	0.879	0.870	0.931	0.916	0.792	0.781
BP	0.856	0.849	0.919	0.892	0.785	0.776