Combining Sub-symbolic and Symbolic Methods for Explainability

Similarly to other connectionist models, Graph Neural Networks (GNNs) lack transparency in their decision-making. A number of sub-symbolic approaches have been developed provide insights into the GNN decision making process. These are first important steps on way explainability, but generated explanations often hard understand for users that not AI experts. To overcome this problem, we introduce a conceptual approach combining and symbolic methods human-centric explanations, incorporate domain knowledge causality. We furthermore notion fidelity as metric evaluating how close explanation is GNN's internal The evaluation with chemical dataset ontology shows explanatory value reliability our method.