The use of class imbalanced learning methods on ULSAM data to predict the case–control status in genome-wide association studies

Journal of Big Data

Table 7 The performances of random forest with imbalanced learning methods with using clumped SNPs

Imbalanced learning methods number of SNP 29		Prediction class		PPV*	NPV**	Sensitivity	Specificity	F1 score	Accuracy
RF		Controls	Cases
Reel class	Controls	278	0	0.00	0.82	0.00	1.00	0.00	0.82
Reel class	Cases	57	0	0.00	0.82	0.00	1.00	0.00	0.82
SMOTE		Controls	Cases
Reel class	Controls	184	90	0.75	0.84	0.89	0.67	0.81	0.78
Reel class	Cases	34	263	0.75	0.84	0.89	0.67	0.81	0.78
SVM SMOTE
Reel class	Controls	260	18	0.90	0.65	0.53	0.94	0.67	0.73
Reel class	Cases	138	155	0.90	0.65	0.53	0.94	0.67	0.73
ADASYN		Controls	Cases
Reel class	Controls	216	68	0.76	0.74	0.74	0.76	0.75	0.75
Reel class	Cases	76	215	0.76	0.74	0.74	0.76	0.75	0.75
RUS		Controls	Cases
Reel class	Controls	39	17	0.67	0.83	0.81	0.70	0.74	0.75
Reel class	Cases	8	35	0.67	0.83	0.81	0.70	0.74	0.75