Table 1.

Performance of different manifold learning methods on the test set

Method	SEN	SPE	PRE	F1	ACC	MCC	AUC
LLE (10)	0.653	0.711	0.607	0.629	0.687	0.361	0.693
ISOMAP (10)	0.687	0.766	0.692	0.695	0.709	0.476	0.738
SLLE (3)	0.671	0.732	0.648	0.656	0.691	0.381	0.703
S-ISOMAP (3)	0.707	0.819	0.721	0.713	0.768	0.508	0.773

The highest value in each column is shown in bold. The numbers in parentheses represent the feature dimensions after dimensionality reduction