Related Experiment Video
Updated: Nov 10, 2025
11:15
Applications of EEG Neuroimaging Data: Event-related Potentials, Spectral Power, and Multiscale Entropy
Published on: June 27, 2013
34.1K
EEG-Based Emotion Recognition Using an Improved Weighted Horizontal Visibility Graph.
Tianjiao Kong1,2, Jie Shao1,2, Jiuyuan Hu1,2
1College of Electronic and Information Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China.
Sensors (Basel, Switzerland)
|April 3, 2021
Summary
This study introduces a novel complex network approach using weighted horizontal visibility graphs for emotion recognition from electroencephalogram (EEG) signals. The method achieves high accuracy in classifying emotions based on EEG data.
Area of Science:
- Neuroscience
- Computer Science
- Signal Processing
Background:
- Emotion recognition is a challenging research area with growing interest.
- Electroencephalogram (EEG) signals offer a promising avenue for objective emotion assessment.
- Extracting meaningful features from complex EEG data remains a key challenge.
Purpose of the Study:
- To develop a novel method for emotion recognition using complex network features derived from EEG signals.
- To introduce Forward Weighted Horizontal Visibility Graphs (FWHVG) and Backward Weighted Horizontal Visibility Graphs (BWHVG) for EEG analysis.
- To evaluate the effectiveness of these complex network features in classifying emotional states.
Main Methods:
- Constructed FWHVG and BWHVG based on angle measurement from EEG signals.
- Extracted network features from the constructed graphs.
- Fused the network feature matrices.
- Classified EEG signals using the fused feature matrix, with and without time-domain features.
Main Results:
- The proposed method achieved average accuracies of 97.53% for valence and 97.75% for arousal using complex network features alone.
- When combined with time-domain features, the classification accuracies reached 98.12% for valence and 98.06% for arousal.
- The novel graph-based approach demonstrated high efficacy in EEG-based emotion recognition.
Conclusions:
- The proposed FWHVG and BWHVG methods provide effective complex network features for EEG-based emotion recognition.
- Feature fusion significantly enhances classification performance.
- This approach offers a robust and accurate method for understanding emotional states from neural signals.
