Causality-based CTR prediction using graph neural networks

•Propose a causality-based CTR prediction model in the GNNs framework (Causal-GNN) integrating representations of feature graph, user graph and ad graph.•Design a structured representation learning method (GraphFwFM) to capture high-order representations on feature graph based on causal inference.•Conduct experiments on three public datasets to show the superiority of Causal-GNN by comparing with several state-of-the-art baselines.