Utilizing Description Logics for Global Explanations of Heterogeneous Graph Neural Networks

Open in new window