Table 6 Performance results for the sensitivity, specificity, accuracy, and F1 score of the features selected according to the five feature selection methods.
From: Combining meta and ensemble learning to classify EEG for seizure detection
Type | Feature | Sen(%) | Spe(%) | F1(%) | Acc(%) | |
|---|---|---|---|---|---|---|
mixture | θ-13,δ-29,θ-29,θ-21, α-13,δ-21 | 90.38 | 91.74 | 72.78 | 91.57 | |
Semi-JMI | θ-13,γ-14,δ-29, θ-29,θ-21,δ-28 | 92.58 | 92.51 | 75.54 | 92.52 | |
JMI | θ-13,γ-14, θ-29,δ-29,θ-21,δ-28 | 91.95 | 92.95 | 76.19 | 92.83 | |
Semi-MIM | θ-13,θ-29,α-13,δ-29,θ-21,α-29 | 90.27 | 90.84 | 70.92 | 90.77 | |
MIM | θ-13,θ-29,α-13,δ-29,θ-21,δ-21 | 90.74 | 91.72 | 72.93 | 91.60 | |
