Heterogeneous Graph Neural Network

TLDR

Representation learning on heterogeneous graphs seeks node embeddings that support tasks such as link prediction and recommendation, yet existing methods struggle to jointly capture diverse structural relations and node attributes. This work introduces HetGNN, a heterogeneous graph neural network designed to jointly model structural and attribute heterogeneity. HetGNN samples strongly correlated heterogeneous neighbors via random walk with restart, groups them by node type, and processes them through a two‑module network that first learns deep content interactions to produce node‑specific embeddings and then aggregates these across neighbor groups while weighting their influence, all trained end‑to‑end with a graph context loss. Experiments on multiple datasets show HetGNN surpasses state‑of‑the‑art baselines in link prediction, recommendation, node classification, clustering, and inductive node classification and clustering.

Abstract

Representation learning in heterogeneous graphs aims to pursue a meaningful vector representation for each node so as to facilitate downstream applications such as link prediction, personalized recommendation, node classification, etc. This task, however, is challenging not only because of the demand to incorporate heterogeneous structural (graph) information consisting of multiple types of nodes and edges, but also due to the need for considering heterogeneous attributes or contents (e.g., text or image) associated with each node. Despite a substantial amount of effort has been made to homogeneous (or heterogeneous) graph embedding, attributed graph embedding as well as graph neural networks, few of them can jointly consider heterogeneous structural (graph) information as well as heterogeneous contents information of each node effectively. In this paper, we propose HetGNN, a heterogeneous graph neural network model, to resolve this issue. Specifically, we first introduce a random walk with restart strategy to sample a fixed size of strongly correlated heterogeneous neighbors for each node and group them based upon node types. Next, we design a neural network architecture with two modules to aggregate feature information of those sampled neighboring nodes. The first module encodes "deep" feature interactions of heterogeneous contents and generates content embedding for each node. The second module aggregates content (attribute) embeddings of different neighboring groups (types) and further combines them by considering the impacts of different groups to obtain the ultimate node embedding. Finally, we leverage a graph context loss and a mini-batch gradient descent procedure to train the model in an end-to-end manner. Extensive experiments on several datasets demonstrate that HetGNN can outperform state-of-the-art baselines in various graph mining tasks, i.e., link prediction, recommendation, node classification & clustering and inductive node classification & clustering.