Table 4.

Performance of different machine learning methods and deep learning methods on Dset_448, Dset_72, and Dset_164.

	TPR	TNR	Pre	ACC	F1	MCC	AUROC	AP
Dset_448
XGBoost	0.477	0.886	0.396	0.830	0.433	0.336	0.792	0.409
LGBM	0.524	0.865	0.378	0.819	0.439	0.341	0.797	0.415
CatBoost	0.498	0.869	0.374	0.819	0.427	0.326	0.786	0.398
SGD	0.480	0.900	0.431	0.843	0.454	0.364	0.799	0.430
MLP	0.480	0.900	0.431	0.843	0.454	0.364	0.799	0.430
TextCNN	0.478	0.893	0.411	0.836	0.442	0.348	0.797	0.419
Capsule	0.537	0.871	0.395	0.826	0.455	0.360	0.806	0.422
GRU	0.534	0.881	0.414	0.834	0.466	0.374	0.815	0.448
LSTM	0.540	0.881	0.416	0.835	0.470	0.378	0.815	0.449
MultiHead	0.524	0.888	0.423	0.838	0.468	0.377	0.815	0.448
EDLMPPI	0.452	0.922	0.477	0.858	0.464	0.383	0.820	0.460
Dset_72
XGBoost	0.501	0.818	0.246	0.785	0.330	0.239	0.743	0.274
LGBM	0.410	0.880	0.288	0.830	0.338	0.249	0.759	0.295
CatBoost	0.406	0.876	0.279	0.826	0.330	0.240	0.751	0.270
SGD	0.521	0.812	0.247	0.781	0.335	0.246	0.748	0.293
MLP	0.504	0.826	0.256	0.792	0.340	0.250	0.759	0.283
TextCNN	0.418	0.878	0.289	0.829	0.342	0.253	0.760	0.301
Capsule	0.544	0.829	0.274	0.799	0.364	0.282	0.773	0.305
GRU	0.425	0.878	0.293	0.830	0.347	0.259	0.774	0.308
LSTM	0.421	0.881	0.296	0.832	0.347	0.260	0.765	0.316
MultiHead	0.279	0.940	0.354	0.870	0.312	0.244	0.776	0.308
EDLMPPI	0.318	0.938	0.377	0.872	0.345	0.276	0.788	0.330
Dset_164
XGBoost	0.553	0.759	0.336	0.722	0.419	0.263	0.723	0.361
LGBM	0.480	0.821	0.372	0.759	0.419	0.274	0.733	0.375
CatBoost	0.576	0.741	0.330	0.711	0.419	0.263	0.719	0.364
SGD	0.496	0.804	0.359	0.749	0.417	0.268	0.730	0.371
MLP	0.457	0.822	0.362	0.756	0.404	0.255	0.720	0.358
TextCNN	0.412	0.868	0.409	0.786	0.410	0.280	0.728	0.394
Capsule	0.438	0.852	0.395	0.777	0.415	0.279	0.741	0.388
GRU	0.443	0.853	0.399	0.779	0.420	0.285	0.747	0.394
LSTM	0.447	0.855	0.405	0.781	0.425	0.291	0.751	0.401
MultiHead	0.447	0.855	0.405	0.781	0.425	0.291	0.751	0.401
EDLMPPI	0.323	0.916	0.460	0.809	0.380	0.277	0.755	0.413