
The core message of this article is that a novel method using Bayesian structure learning (BSL) to estimate dynamic functional connectivity of EEG signals, combined with a graph convolutional network (GCN) classifier, can effectively classify working memory load with high accuracy, outperforming state-of-the-art methods.


coremsg

bayesian-functional-connectivity-and-graph-convolutional-network-for-classifying-working-memory-load


Bayesian Functional Connectivity and Graph Convolutional Network for Classifying Working Memory Load
