Enhancing Explainability of Graph Neural Networks Through Conceptual and Structural Analyses and Their Extensions

TL;DR

This paper proposes a novel explainability framework for Graph Neural Networks that enhances understanding of their decision processes by analyzing both conceptual and structural aspects, aiming for more efficient and adaptable explanations.

Contribution

It introduces a new XAI framework for GNNs that combines structural and conceptual analyses to improve interpretability and computational efficiency.

Findings

Framework provides more accurate explanations of GNN decisions.

Reduces computational resources needed for explanations.

Improves generalizability of explanations across scenarios.

Abstract

Graph Neural Networks (GNNs) have become a powerful tool for modeling and analyzing data with graph structures. The wide adoption in numerous applications underscores the value of these models. However, the complexity of these methods often impedes understanding their decision-making processes. Current Explainable AI (XAI) methods struggle to untangle the intricate relationships and interactions within graphs. Several methods have tried to bridge this gap via a post-hoc approach or self-interpretable design. Most of them focus on graph structure analysis to determine essential patterns that correlate with prediction outcomes. While post-hoc explanation methods are adaptable, they require extra computational resources and may be less reliable due to limited access to the model's internal workings. Conversely, Interpretable models can provide immediate explanations, but their generalizability to different scenarios remains a major concern. To address these shortcomings, this thesis seeks to develop a novel XAI framework tailored for graph-based machine learning. The proposed framework aims to offer adaptable, computationally efficient explanations for GNNs, moving beyond individual feature analysis to capture how graph structure influences predictions.