Table 4.
Comparison between the proposed method and related works.
| Method | Number of classes | Feature extraction algorithm | Number of electrodes | Classifiers | AVG accuracy(%) |
|---|---|---|---|---|---|
| 18 | 3 emotions (anger, surprise, other) | Minimum redundancy maximum relevance (mRMR) | 32 | SVM-random forest | 60 |
| 14 | 4 emotions (happy, sad, angry, and relaxed) | Time and frequency domain features | 5 | Decision tree algorithm | 81.64 |
| 15 | 4 emotions (angry, sad, happy, and relaxed) | time domain features, frequency domain features and entropy | 32 | ANN | 93.75 |
| 13 | 4 emotions | Probability distribution for wavelet packet coefficient | 3 | SVM | 70.5 |
| 19 | 9 emotions | Spectral features | 32 | DBN | 79.2 |
| 8 | 9 emotions | Fusing of 6 statistical features | 32 | SVM | 81.87 |
| Proposed method | 9 emotions | ZTWBES | Adaptive | QDC | 87.05 |
| RNN-scheme 1 | 89.33 | ||||
| RNN-scheme 2 | 86.53 |