Improvement of Roller Bearing Diagnosis with Unlabeled Data Using Cut Edge Weight Confidence Based Tritraining

Table 6

Classification error rate of initial and final hypothesis and the corresponding improvements of self-training under different unlabeling rate.

Case	Unlabeling rate of 0.8			Unlabeling rate of 0.6			Unlabeling rate of 0.4			Unlabeling rate of 0.2
	Initial	Final	Improvement	Initial	Final	Improvement	Initial	Final	Improvement	Initial	Final	Improvement
Case 1	0.3353	0.2933	12.53%	0.2168	0.2033	6.23%	0.1002	0.0933	6.89%	0.0417	0.0454	−8.15%
Case 2	0.4737	0.4058	14.33%	0.4037	0.3658	9.39%	0.3158	0.2632	16.66%	0.2811	0.2648	5.80%