Abstract
Graph Neural Networks (GNNs) have achieved remarkable success in node classification. Building on this progress, Heterogeneous Graph Neural Networks (HGNNs) integrate relation types and node and edge semantics to leverage heterogeneous information. However, whether HGNNs are intrinsically effective for node classification remains underexamined, and most studies implicitly assume rather than establish this effectiveness.
In this work, we examine HGNNs for node classification from two perspectives: model architecture and heterogeneous information. We conduct a systematic reproduction across 21 datasets and 20 baselines, complemented by comprehensive hyperparameter retuning.
To further disentangle the source of performance gains, we develop a causal mediation analysis framework that treats the introduction of heterogeneous relation information as the treatment, candidate structural properties as mediators, and node classification performance as the outcome. This framework first screens candidate mediators according to their treatment-induced changes and their associations with performance improvement, and then decomposes the total effect into mediated and direct effects.
Our results lead to two conclusions. First, model architecture and complexity have no causal effect on node classification performance. Second, heterogeneous information exerts a positive causal effect primarily through increasing homophily and local-global distribution discrepancy, which makes node classes more distinguishable.
The implementation is publicly available at CausalHGNN GitHub.
Blogger's Review: This article employs a causal analysis approach to delve into the role of HGNNs in node classification, revealing the different impacts of model architecture and heterogeneous information on performance. The innovative methodology provides valuable insights for future research, particularly in the application potential of causal inference.