Work

EEG-UDAF: A novel unsupervised domain adaptation framework leveraging dual-view frequency and spatio-temporal features for EEG-based emotion recognition

Year: 2026

Type: article

Abstract: Currently, machine learning technology has been widely used in electroencephalogram (EEG) automatic emotion recognition tasks. However, the non-stationarity, individual differences, and high annotation costs in EEG signals seriously affect the generalization ability of the model. To solve these problems, we propose a new unsupervised domain adaptation framework named EEG-UDAF for EEG emotion recognition. First, we design a dual-view frequency domain feature fusion module that can extract rich EEG signals frequency domain features from two perspectives, differential entropy and power spectral density, and fuse them based on the self-attention mechanism. Second, we propose a spatio-temporal feature extraction module based on graph convolutional network and bidirectional long short-term memory. It can comprehensively consider the spatial topological structure of the brain and the dynamic characteristics of temporal changes between EEG channels to extract complex spatio-temporal features. Third, we use the unsupervised domain adaptation strategy to solve the non-stationarity and individual differences in EEG signals without manual annotation, and introduce a semantic correction mechanism to prevent incorrect class alignment. Experiments based on intra-subject and inter-subject paradigms show that EEG-UDAF can effectively alleviate non-stationarity and individual differences. Moreover, these transferable EEG representations also provide a potential foundation for LLM-augmented multimodal data fusion in large-scale affective computing.

Source: Journal Of Big Data

DOI: 10.1186/s40537-026-01584-5