Utilizing Description Logics for Global Explanations of Heterogeneous Graph Neural Networks
This addresses the need for better interpretability in GNNs for researchers and practitioners, though it is incremental as it builds on existing explanation methods by incorporating description logic.
The paper tackled the problem of limited explainability in Graph Neural Networks (GNNs) for node classification by proposing the use of description logic class expressions to provide more expressive global explanations for heterogeneous graphs, resulting in a method that identifies the best explanation using scoring functions based on aggregated predictions or fidelity comparisons.
Graph Neural Networks (GNNs) are effective for node classification in graph-structured data, but they lack explainability, especially at the global level. Current research mainly utilizes subgraphs of the input as local explanations or generates new graphs as global explanations. However, these graph-based methods are limited in their ability to explain classes with multiple sufficient explanations. To provide more expressive explanations, we propose utilizing class expressions (CEs) from the field of description logic (DL). Our approach explains heterogeneous graphs with different types of nodes using CEs in the EL description logic. To identify the best explanation among multiple candidate explanations, we employ and compare two different scoring functions: (1) For a given CE, we construct multiple graphs, have the GNN make a prediction for each graph, and aggregate the predicted scores. (2) We score the CE in terms of fidelity, i.e., we compare the predictions of the GNN to the predictions by the CE on a separate validation set. Instead of subgraph-based explanations, we offer CE-based explanations.