An automatic emotion recognition method using discrete wavelet transform, tabu search algorithm, and a fisher-KNN classifier successfully classified joy and sadness from ECG signals in 391 subjects.
The study demonstrates the feasibility of using discrete wavelet transform, tabu search, and a fisher-KNN classifier for emotion recognition from ECG signals.
Emotion recognition based on physiological signals is an important research fields with promising application future. This paper firstly carried out the work of affective (joy and sadness) electrocardiogram (ECG) signal acquisition obtained from 391 subjects through stimulation of film clips. The automatic location of P-QRS-T wave was performed by use of discrete wavelet transform (DWT), which was crucial for ECG feature extraction by the computer. Through tabu search algorithm (TS), the best combination of the ECG emotion features was selected for classification. Finally fisher-KNN proposed in this paper was implemented to classify the test data. An effective emotion feature subset and a better recognition result were achieved availably. This research showed the feasibility of the method which sought the affective ECG features. And it was practicable to apply TS and fisher-KNN classifier for emotion recognition based on ECG signal.
Cai et al. (2009) studied Emotion recognition (n=391). Emotion recognition using discrete wavelet transform, tabu search algorithm, and fisher-KNN classifier was evaluated on Classification of joy and sadness from ECG signals. An automatic emotion recognition method using discrete wavelet transform, tabu search algorithm, and a fisher-KNN classifier successfully classified joy and sadness from ECG signals in 391 subjects.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: