Deep learning-based classification model for GPR151 activator activity prediction

BMC Bioinformatics

Table 4 Performance comparision of the test set on five traditional classifiers, and the bold marks the best in the group

Feature (raw)	Classifier	Accuracy	Precision	Recall	F1
RDKFP	LR	0.8643	0.6462	0.5638	0.6022
	KNN	0.8658	0.7462	0.3992	0.5201
	RF	0.8651	0.8	0.3457	0.4828
	DT	0.8043	0.4619	0.4486	0.4551
	SVM	0.8898	0.8478	0.4815	0.6142
MorganFP	LR	0.8928	0.7404	0.6377	0.6829
	KNN	0.8598	0.8043	0.3045	0.4418
	RF	0.8741	0.8378	0.3827	0.5254
	DT	0.8568	0.6313	0.5144	0.5669
	SVM	0.9018	0.8256	0.5844	0.6843
Mol2vec	LR	0.8928	0.766	0.5926	0.6682
	KNN	0.8921	0.7346	0.6379	0.6828
	RF	0.8838	0.775	0.5103	0.6154
	DT	0.8313	0.5375	0.5309	0.5342
	SVM	0.8898	0.7727	0.5597	0.6492

ISSN: 1471-2105