Improving the Quality of Rule-Based GNN Explanations
Abstract
Graph Neural Networks (GNNs) have gained significant attention for their ability to handle graph-structured data. However, the lack of transparency in their decision-making process has led to increased interest in explainable AI for GNNs. Rule-based explanations provide interpretable insights into GNN predictions, but their quality can vary significantly. This paper addresses the challenge of improving the quality of rule-based explanations for GNN models through novel methodological approaches and evaluation metrics.
Citation
@inproceedings{kamal2022improving,
title={Improving the Quality of Rule-Based GNN Explanations},
author={Kamal, Ataollah and Vincent, Elouan and Plantevit, Marc and Robardet, Celine},
note={accepted},
booktitle={Workshop on eXplainable Knowledge Discovery in Data Mining. Machine Learning and Principles and Practice of Knowledge Discovery in Databases - International Workshops of ECML PKDD 2022, Grenoble, France, September 19-23, 2022, Proceedings, Part I},
address={Grenoble, France},
pages={467--482},
year={2022},
month={Sep},
doi={10.1007/978-3-031-23618-1\_31},
pdf={https://hal.science/hal-04580342v1/file/kamal.22.xkdd.pdf},
hal_id={hal-04580342},
hal_version={v1}
}
Kamal, A., Vincent, E., Plantevit, M., & Robardet, C. (2022). Improving the Quality of Rule-Based GNN Explanations. In Workshop on eXplainable Knowledge Discovery in Data Mining, ECML PKDD 2022 (pp. 467-482). Grenoble, France.
Kamal, Ataollah, et al. "Improving the Quality of Rule-Based GNN Explanations." Workshop on eXplainable Knowledge Discovery in Data Mining, ECML PKDD 2022. Grenoble, France, 2022. 467-482.