BMC Bioinformatics

Table 3 Predictive performance metrics of our models

From: Prediction of HIV-1 protease cleavage site from octapeptide sequence information using selected classifiers and hybrid descriptors

Model	Performance on testing set						10-Fold CV
Model	B. Acc	AUC	Spec	Sen	F-Index	JI	B. Acc	AUC	Spec	Sen	F-Index	JI
LSVM	92.8	0.97	0.93	0.92	0.88	0.80	79.2	0.96	0.94	0.64	0.80	0.70
LDA	88.4	0.96	0.96	0.80	0.88	0.80	84.7	0.95	0.95	0.74	0.85	0.75
GBC	89.8	0.97	0.97	0.83	0.90	0.82	80.1	0.91	0.95	0.65	0.81	0.71
KNN	69.0	0.78	0.91	0.47	0.69	0.57	68.5*	0.77	0.91	0.46	0.69	0.57
NBC	87.0	0.94	0.97	0.77	0.88	0.79	83.0	0.90	0.92	0.74	0.84	0.77
DTC	73.6	0.75	0.90	0.57	0.73	0.59	75.9	0.76	0.92	0.60	0.75	0.64
MLPC	92.0	0.97	0.95	0.89	0.90	0.82	87.5	0.96	0.95	0.75	0.86	0.78
PC	50.3	0.89	1.00	0.00	0.46	0.42	60.0*	0.82	0.85	0.35	0.53	0.46
LR	90.9	0.97	0.97	0.84	0.91	0.83	86.4	0.96	0.97	0.76	0.88	0.79

Values represented in the 10-Fold Experiment table are the averages across the 10-folds; *Significant at p < 0.05

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com