EEG-based Graph-guided Domain Adaptation for Robust Cross-Session Emotion Recognition

TL;DR

This paper introduces EGDA, a graph-guided domain adaptation framework that enhances cross-session EEG-based emotion recognition by aligning data distributions and preserving EEG structure, achieving state-of-the-art accuracy.

Contribution

EGDA is the first method to jointly align global and class-specific EEG distributions with graph regularization for improved emotion recognition across sessions.

Findings

EGDA achieves over 80% accuracy on SEED-IV transfer tasks.

Gamma band is most discriminative for emotion recognition.

Central-parietal and prefrontal regions are key brain areas identified.

Abstract

Accurate recognition of human emotional states is critical for effective human-machine interaction. Electroencephalography (EEG) offers a reliable source for emotion recognition due to its high temporal resolution and its direct reflection of neural activity. Nevertheless, variations across recording sessions present a major challenge for model generalization. To address this issue, we propose EGDA, a framework that reduces cross-session discrepancies by jointly aligning the global (marginal) and class-specific (conditional) distributions, while preserving the intrinsic structure of EEG data through graph regularization. Experimental results on the SEED-IV dataset demonstrate that EGDA achieves robust cross-session performance, obtaining accuracies of 81.22%, 80.15%, and 83.27% across three transfer tasks, and surpassing several baseline methods. Furthermore, the analysis highlights the Gamma frequency band as the most discriminative and identifies the central-parietal and prefrontal brain regions as critical for reliable emotion recognition.