BMC Medical Informatics and Decision Making

Table 2 Comparison results on the datasets of binary classification tasks

From: MODILM: towards better complex diseases classification using a novel multi-omics data integration learning model

Method	ROSMAP dataset			LGG-2 dataset
Method	ACC	F1	AUC	ACC	F1	AUC
KNN	0.657	0.671	0.709	0.729	0.738	0.799
SVM	0.770	0.778	0.770	0.737	0.748	0.810
RF	0.726	0.734	0.811	0.756	0.767	0.840
block PLSDA	0.742	0.755	0.830	0.729	0.738	0.799
NN	0.755	0.764	0.826	0.754	0.757	0.754
XGBoost	0.760	0.791	0.837	0.748	0.742	0.823
DeepMO	0.772	0.780	0.801	0.765	0.760	0.786
CDForest	0.778	0.791	0.839	0.843	0.858	0.871
P-NET	0.805	0.810	0.818	0.886	0.890	0.897
MOMA	0.818	0.826	0.875	0.942	0.939	0.950
MOGONET	0.815	0.821	0.874	0.951	0.958	0.961
Our MODILM	0.843	0.850	0.891	0.975	0.978	0.993

Back to article page

ISSN: 1472-6947

Contact us

General enquiries: journalsubmissions@springernature.com