Improved Pre-miRNA Classification by Reducing the Effect of Class Imbalance

Table 5

The classification results of MiRNAClassify and other methods over the merged datasets and the updated datasets.

Method	Accuracy (%)	Merged human dataset	Merged animal dataset	Merged plant dataset	Updated human dataset	Updated ath dataset	Updated osa dataset	Updated gma dataset
MiRNAClassify	SE	97.93	95.85	93.37
	SP	98.30	97.62	97.91	94.08	92.65	79.29	93.05
		98.11	96.73	95.61
MicroPred	SE	92.25	91.61	89.50
	SP	95.70	94.85	93.10	91.27	89.71	66.86	85.76
		93.96	93.21	91.28
PlantMiRNAPred	SE	93.58	92.70	91.39
	SP	91.20	88.60	87.10	92.11	89.71	70.42	88.41
		92.38	90.63	89.22
HuntMi	SE	95.32	94.14	91.76
	SP	97.11	96.07	95.94	92.68	91.18	72.78	89.07
		96.21	95.10	93.83
miRNApre	SE	97.39	93.49	91.71
	SP	90.90	89.80	88.10	90.14	89.71	71.01	86.09
		94.09	91.63	89.89