MAGNN: Metapath Aggregated Graph Neural Network for Heterogeneous Graph Embedding

By Xinyu Fu, Jiani Zhang, Ziqiao Meng, Irwin King. The Web Conference, 2020

1,182 citations recorded by the source corpus.

A large number of real-world graphs or networks are inherently heterogeneous, involving a diversity of node types and relation types. Heterogeneous graph embedding is to embed rich structural and semantic information of a heterogeneous graph into low-dimensional node representations. Existing models usually define multiple metapaths in a heterogeneous graph to capture the composite relations and guide neighbor selection. However, these models either omit node content features, discard intermediate nodes along the metapath, or only consider one metapath. To address these three limitations, we propose a new model named Metapath Aggregated Graph Neural Network (MAGNN) to boost the final performance. Specifically, MAGNN employs three major components, i.e., the node content transformation to encapsulate input node attributes, the intra-metapath aggregation to incorporate intermediate semantic nodes, and the inter-metapath aggregation to combine messages from multiple metapaths. Extensive experiments on three real-world heterogeneous graph datasets for node classification, node clustering, and link prediction show that MAGNN achieves more accurate prediction results than state-of-the-art baselines.

Subfields: Artificial Intelligence, Molecular Biology, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/c7fd29fdd2e0b50a571db4f607eab138e9ecb644

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Philip S. Yu
2. Chuan Shi
3. Xiao Wang
4. Jie Tang
5. Quanming Yao
6. Tong Zhang
7. Yuxiao Dong
8. Yuhui Ding

In Molecular Biology

1. Chuan Shi
2. Xiao Wang
3. Yuhui Ding
4. Quanming Yao
5. Tong Zhang
6. Yaoman Li
7. Philip S. Yu
8. Chang Zhou

In Statistical and Nonlinear Physics

1. Chuan Shi
2. Xiao Wang
3. Quanming Yao
4. Tong Zhang
5. Yuhui Ding
6. Jie Tang
7. Yuxiao Dong
8. Binbin Hu

Citations in the corpus

• Heterogeneous Graph Structure Learning for Graph Neural Networks (2021, methodology): “However, this method not only costs large memory with the computation of stacking multiple layers of adjacency matrices, but also discards the intermediate nodes which leads to information loss (Fu et al. 2020).”
• AutoGraph: Automated Graph Neural Network (2020, methodology): “Inspired by CNNs [10,9] and graph embedding [2,4], GNNs are proposed to collectively aggregate information from graph structure.”
• DiffMG: Differentiable Meta Graph Search for Heterogeneous Graph Neural Networks (2020, methodology): “where X ′ denotes the projected features output by type-specific transformation [7, 26] which projects features of different node Research Track Paper KDD ’21, August 14–18, 2021, Virtual Event, Singapore”
• A Survey on Heterogeneous Graph Embedding: Methods, Techniques, Applications and Sources (2020, methodology): “Recently, HGNNs have attracted considerable attention [15], [16], [74], [72].”
• Self-supervised Heterogeneous Graph Neural Network with Co-contrastive Learning (2021, methodology): “A.1 Baselines The publicly available implementations of baselines can be found at the following URLs: • GraphSAGE: https://github.com/williamleif/GraphSAGE • GAE: https://github.com/tkipf/gae • Mp2vec: https://ericdongyx.github.io/metapath2vec/m2v.html • HERec: https://github.com/librahu/HERec • Het”
• LLMRec: Large Language Models with Graph Augmentation for Recommendation (2023, methodology): “We adjust feature vectors using the aggregation weight 𝜔 1 and 𝐿 2 normalization to mitigate distribution gaps [9], ensurring the effectiveness of additional features within the recommender encoder.”
• Are we really making much progress?: Revisiting, benchmarking and refining heterogeneous graph neural networks (2021, methodology): “Take node classification for example, numerous HGNNs, such as HAN [36], GTN [43], RSHN [45], HetGNN [44], MAGNN [12], HGT [20], and HetSANN [17] were developed within the last two years.”
• A Comprehensive Survey on Community Detection With Deep Learning (2021, methodology): “In addition, community structures in complicated real-world scenarios, such as large scale [18], high-sparse [75], complex structural [76], [77], and dynamic [78] networks, can be better explored benefiting from the capacity of deep learning to handle big data.”
• A Comprehensive Survey on Graph Anomaly Detection With Deep Learning (2021, background): “Opportunities: Although plenty of deep learning techniques have been developed to analyze heterogeneous graphs [151], [152], incorporating the most recent techniques for graph anomaly detection is uncharted yet.”
• Propagation Model Search for Graph Neural Networks (2020, methodology): “GNNs in Section 2.1 are intended for homogeneous graphs, but they do not take into account that many real-world problems come with heterogeneous graphs which contain multiple types of nodes and edges (Sun et al. 2011).”