Competing Endogenous RNA and Coexpression Network Analysis for Identification of Potential Biomarkers and Therapeutics in association with Metastasis Risk and Progression of Prostate Cancer

Table 1

Performance of classification models by 5-fold cross-validation, random sampling, and leave-one-out validation.

Method	5-fold cross-validation		Random sampling		Leave-one-out
	AUC	Precise	AUC	Precise	AUC	Precise
Logistic regression (LR)	0.442	0.683	0.528	0.640	0.459	0.706
Random forest (RF)	0.840	0.842	0.815	0.844	0.775	0.840
nearest neighbors (kNN)	0.581	0.712	0.648	0.753	0.571	0.719
Naive Bayesian (NB)	0.772	0.821	0.762	0.810	0.760	0.815