EEG emotion recognition based on differential entropy feature matrix through 2D-CNN-LSTM network

EURASIP Journal on Advances in Signal Processing

Table 3 Two-dimensional-CNN-LSTM classification result

Subject	Binary classification of valence			Binary classification of arousal
Subject	Accuracy (%)	F-score (%)	Kappa (%)	Accuracy (%)	F-score (%)	Kappa (%)
01	92.16	89.73	88.89	90.23	90.36	91.33
02	90.14	90.56	91.02	90.58	91.22	91.27
03	91.33	89.77	90.59	91.36	90.88	92.47
04	90.65	90.25	91.76	91.27	91.88	88.87
05	91.58	90.27	91.01	89.77	90.28	89.67
06	89.89	90.21	91.02	92.11	91.29	92.02
07	91.35	91.89	92.03	90.23	91.22	90.78
08	91.54	92.32	92.01	86.54	89.67	88.73
09	91.39	92.86	90.89	91.36	92.24	91.69
10	91.11	92.25	92.42	91.50	93.51	92.61
11	91.58	92.19	92.54	91.54	90.32	91.29
12	90.51	91.40	91.12	90.38	90.88	90.71
13	91.15	92.89	91.25	90.33	90.58	91.58
14	89.91	92.43	90.78	92.38	90.05	91.11
15	92.63	91.97	92.18	92.56	91.78	92.82
16	90.91	91.88	91.61	92.24	91.89	92.99
17	91.58	91.54	92.56	92.69	91.38	91.72
18	91.56	92.69	92.65	92.58	93.11	92.89
19	91.67	91.91	92.68	90.78	91.32	91.96
20	90.94	91.37	91.82	91.84	92.77	92.65
21	93.01	91.56	92.22	92.49	91.88	91.84
22	89.68	90.28	92.28	91.64	91.56	91.77
23	91.37	91.36	91.28	91.21	91.85	91.29
24	89.89	91.76	92.88	91.59	90.68	91.21
25	90.16	91.58	90.36	90.71	91.36	92.21
26	90.83	92.49	91.34	90.88	90.21	91.06
27	91.56	90.99	91.56	89.78	89.88	90.09
28	90.81	91.21	92.02	90.36	91.26	91.28
29	92.02	91.25	92.45	91.03	91.45	91.17
30	91.06	90.85	91.49	90.88	90.87	92.85
31	91.33	91.54	92.55	91.02	90.65	88.68
32	90.88	91.09	92.35	91.36	90.88	89.75
Average	91.92	91.85	92.36	92.31	90.75	91.76