Table 8.
Ternary sentiment analysis results of the multi-modal model using the convolutional EEG decoding component.
| Randomly initialized | GloVe | BERT | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Model | P | R | F1 (std) | P | R | F1 (std) | P | R | F1 (std) |
| Baseline | 0.408 | 0.384 | 0.351 (0.07) | 0.510 | 0.507 | 0.496 (0.06) | 0.722 | 0.714 | 0.710 (0.05) |
| + noise | 0.359 | 0.388 | 0.334 (0.09) | 0.494 | 0.484 | 0.476 (0.07) | 0.715 | 0.683 | 0.684 (0.05) |
| + ET | 0.417 | 0.399 | 0.372 (0.05) | 0.509 | 0.512 | 0.500 (0.07) | 0.721 | 0.687 | 0.670 (0.05) |
| + EEG full | 0.365 | 0.384 | 0.333 (0.08) | 0.488 | 0.484 | 0.476 (0.06) | 0.738 | 0.724 | 0.723+ (0.04) |
| + EEG θ | 0.389 | 0.372 | 0.330 (0.06) | 0.511 | 0.495 | 0.477 (0.06) | 0.727 | 0.718 | 0.716+ (0.05) |
| + EEG α | 0.357 | 0.382 | 0.331 (0.11) | 0.534 | 0.525 | 0.515+ (0.06) | 0.732 | 0.715 | 0.713+ (0.04) |
| + EEG β | 0.425 | 0.418 | 0.378 (0.08) | 0.534 | 0.529 | 0.520+ (0.05) | 0.727 | 0.717 | 0.715 (0.04) |
| + EEG γ | 0.404 | 0.406 | 0.360 (0.08) | 0.539 | 0.521 | 0.514 (0.06) | 0.733 | 0.725 | 0.721+ (0.04) |
| +θ+α+β+γ | 0.384 | 0.402 | 0.354 (0.10) | 0.517 | 0.504 | 0.488 (0.05) | 0.733 | 0.717 | 0.715 (0.06) |
We report precision (P), recall (R), F1-score and the standard deviation (std) between five runs. The best results per column are marked in bold. Significance is indicated on the F1-score with asterisks: * denotes p < 0.05 (uncorrected), + denotes p < 0.003 (Bonferroni corrected p-value).