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

Table 3 Comparison results on the datasets of multi-classification tasks

Method	BRCA dataset			SKCM dataset			LGG-4 dataset			LUSC dataset
Method	ACC	F1-weighted	F1-macro	ACC	F1-weighted	F1-macro	ACC	F1-weighted	F1-macro	ACC	F1-weighted	F1-macro
KNN	0.742	0.729	0.682	0.772	0.767	0.736	0.739	0.738	0.741	0.722	0.728	0.689
SVM	0.729	0.702	0.640	0.813	0.812	0.805	0.751	0.750	0.754	0.735	0.732	0.598
RF	0.755	0.733	0.649	0.859	0.857	0.827	0.756	0.742	0.733	0.722	0.838	0.524
Block PLSDA	0.642	0.534	0.369	0.860	0.861	0.830	0.76	0.758	0.772	0.754	0.748	0.751
NN	0.754	0.740	0.668	0.847	0.856	0.862	0.789	0.788	0.786	0.766	0.778	0.781
XGBoost	0.781	0.764	0.701	0.881	0.880	0.863	0.810	0.809	0.798	0.778	0.825	0.741
DeepMO	0.782	0.750	0.723	0.855	0.835	0.837	0.821	0.826	0.835	0.771	0.776	0.780
CDForest	0.789	0.756	0.759	0.862	0.851	0.842	0.878	0.886	0.891	0.778	0.781	0.783
P-NET	0.785	0.776	0.712	0.875	0.861	0.865	0.889	0.897	0.901	0.780	0.791	0.782
MOMA	0.816	0.811	0.790	0.905	0.891	0.886	0.939	0.932	0.926	0.839	0.835	0.810
MOGONET	0.829	0.825	0.774	0.913	0.913	0.912	0.943	0.942	0.927	0.855	0.838	0.799
Our MODILM	0.845	0.840	0.804	0.928	0.927	0.925	0.954	0.954	0.948	0.865	0.855	0.833

ISSN: 1472-6947