Table 7 Dataset 2 long window (\(w=1\,\text{s}\)) features grand average (across all participants) accuracy and F1 score of sLDA classification, shown by types of features and types of TFRs calculated with Shannon entropy
From: Detection of motor imagery based on short-term entropy of time–frequency representations
Accuracy (%) | ||||||||
|---|---|---|---|---|---|---|---|---|
TFR | tfrsp | tfrrsp | tfrgabor | tfrrgab | tfrpwv | tfrrpwv | tfrspwv | tfrrspwv |
Feature type | entropy | entropy | entropy | entropy | entropy | entropy | entropy | entropy |
EE vs EF | 53.14 | 52.9 | 53.36 | 53.69 | 52.34 | 52.26 | 51.86 | 51.7 |
EE vs base | 55.55 | 57.64 | 56.26 | 59.17 | 63.67 | 76.86 | 60.28 | 62.1 |
EF vs base | 55.9 | 58.69 | 57.36 | 58.36 | 63.58 | 76.84 | 59.41 | 61.87 |
F1 score (%) | ||||||||
|---|---|---|---|---|---|---|---|---|
EE vs EF | 53.13 | 53.13 | 53.12 | 53.48 | 52.77 | 51.91 | 51.84 | 51.77 |
EE vs base | 54.95 | 56.24 | 55.69 | 57.5 | 61.91 | 76.11 | 58.57 | 60.73 |
EF vs base | 55.44 | 56.82 | 57.29 | 56.52 | 61.66 | 76.02 | 57.66 | 60.41 |