Zhaoliang Chen: College of Computer and Data Science, Fuzhou University, Fuzhou 350116, China; Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing, Fuzhou University, Fuzhou 350116, China.
Zhihao Wu: College of Computer and Data Science, Fuzhou University, Fuzhou 350116, China; Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing, Fuzhou University, Fuzhou 350116, China.
Luying Zhong: College of Computer and Data Science, Fuzhou University, Fuzhou 350116, China; Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing, Fuzhou University, Fuzhou 350116, China.
Claudia Plant: Faculty of Computer Science, University of Vienna, Vienna 1090, Austria; ds:UniVie, Vienna 1090, Austria.
Shiping Wang: College of Computer and Data Science, Fuzhou University, Fuzhou 350116, China; Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing, Fuzhou University, Fuzhou 350116, China.
Wenzhong Guo: College of Computer and Data Science, Fuzhou University, Fuzhou 350116, China; Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing, Fuzhou University, Fuzhou 350116, China. Electronic address: guowenzhong@fzu.edu.cn.
Heterogeneous graph neural networks play a crucial role in discovering discriminative node embeddings and relations from multi-relational networks. One of the key challenges in heterogeneous graph learning lies in designing learnable meta-paths, which significantly impact the quality of learned embeddings. In this paper, we propose an Attributed Multi-Order Graph Convolutional Network (AMOGCN), which automatically explores meta-paths that involve multi-hop neighbors by aggregating multi-order adjacency matrices. The proposed model first constructs different orders of adjacency matrices from manually designed node connections. Next, AMOGCN fuses these various orders of adjacency matrices to create an intact multi-order adjacency matrix. This process is supervised by the node semantic information, which is extracted from the node homophily evaluated by attributes. Eventually, we employ a one-layer simplifying graph convolutional network with the learned multi-order adjacency matrix, which is equivalent to the cross-hop node information propagation with multi-layer graph neural networks. Substantial experiments reveal that AMOGCN achieves superior semi-supervised classification performance compared with state-of-the-art competitors.
