ECG Multi-Emotion Recognition Based on Heart Rate Variability Signal Features Mining.

Area of Science:

• Physiology
• Psychology
• Computer Science

Background:

• Heart rate variability (HRV) reflects autonomic nervous system function and its link to emotions.
• Current HRV emotion recognition methods often overlook local signal features, limiting information utilization.

Purpose of the Study:

• To propose an effective method for emotion recognition using HRV by enhancing feature extraction.
• To improve the accuracy and efficiency of emotion recognition from physiological signals.

Main Methods:

• Developed the HRV Emotion Recognition (HER) method incorporating Amplitude Level Quantization (ALQ) for feature extraction.
• Utilized Emotion Quantification Analysis (EQA) for assessing emotional arousal.
• Employed logistic regression (LR) for classification of extracted HRV features.

Main Results:

• The proposed HER method achieved an average accuracy rate of 84.3% in emotion recognition.
• The HER method effectively extracts and utilizes local features within HRV signals across different frequency ranges.
• Demonstrated superior performance compared to existing HRV-based emotion recognition techniques.

Conclusions:

• The HER method offers an efficient and accurate approach to emotion recognition using HRV.
• This technique enhances the utilization of local HRV signal information for better emotion detection.
• Provides valuable support for research in psychology, medicine, and affective computing.