EDGE: Evaluation Framework for Logical vs. Subgraph Explanations for Node Classifiers on Knowledge Graphs
R Sapkota, D Köhler, S Heindorf - Proceedings of the 33rd ACM …, 2024 - dl.acm.org
As machine learning and deep learning become increasingly integrated into our daily lives,
understanding how these technologies make decisions is crucial. To ensure transparency …
understanding how these technologies make decisions is crucial. To ensure transparency …
Utilizing Description Logics for Global Explanations of Heterogeneous Graph Neural Networks
D Köhler, S Heindorf - arXiv preprint arXiv:2405.12654, 2024 - arxiv.org
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 …
but they lack explainability, especially at the global level. Current research mainly utilizes …