Motif-based Convolutional Neural Network on Graphs
This work addresses the challenge of applying CNNs to heterogeneous graphs with typed nodes, which is important for domains like social network analysis, but it is incremental as it builds on existing graph CNN approaches.
The paper tackles the problem of generalizing CNNs to irregular and heterogeneous graphs by introducing a spatial convolution operation based on high-order connection patterns (motifs) and an attention model to combine features from multiple patterns, achieving significant gains of 6-21% over existing methods in semi-supervised node classification on real-world datasets.
This paper introduces a generalization of Convolutional Neural Networks (CNNs) to graphs with irregular linkage structures, especially heterogeneous graphs with typed nodes and schemas. We propose a novel spatial convolution operation to model the key properties of local connectivity and translation invariance, using high-order connection patterns or motifs. We develop a novel deep architecture Motif-CNN that employs an attention model to combine the features extracted from multiple patterns, thus effectively capturing high-order structural and feature information. Our experiments on semi-supervised node classification on real-world social networks and multiple representative heterogeneous graph datasets indicate significant gains of 6-21% over existing graph CNNs and other state-of-the-art techniques.