AUTHOR=Tao Jianwen , Dan Yufang 

TITLE=Multi-Source Co-adaptation for EEG-Based Emotion Recognition by Mining Correlation Information

JOURNAL=Frontiers in Neuroscience

VOLUME=Volume 15 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2021.677106

DOI=10.3389/fnins.2021.677106

ISSN=1662-453X

ABSTRACT=Since each individual subject may present completely different encephalogram (EEG) patterns with respect to other subjects, existing subject-independent emotion classifiers trained on data sampled from cross-subjects or cross-dataset generally fail to achieve sound accuracy. For this reason, domain adaptation approaches have been employed to address this issue. Although the effectiveness of multi-source adaptation has recently got extensive attention, there are limited attempts to improve its performance only by using some sharing knowledge among sources. Focusing on automated emotion recognition (AER) by leveraging cross subject or cross dataset EEG features, in this paper we propose a robust Multi-source co-Adaptation framework by mining diverse Correlation Information (MACI) among domains, and features by append   and correlation metric regularization too. Specifically, by minimizing the statistical and semantic distribution differences between the source domain and the target domain, multiple subject-invariant classifiers of different subjects can be learned together in a joint framework, which can make MACI to use the relevant knowledge from multiple sources by exploiting the developed correlation metric function. Comprehensive experimental evidence on DEAP and SEED datasets verifies the better performance of MACI in the EEG-based emotion recognition.