WGCN: Graph Convolutional Networks with Weighted Structural Features

被引:26
作者
Zhao, Yunxiang [1 ]
Qi, Jianzhong [1 ]
Liu, Qingwei [1 ]
Zh, Rui [2 ]
机构
[1] Univ Melbourne, Melbourne, Vic, Australia
[2] Www Ruizhang Info, Melbourne, Vic, Australia
来源
SIGIR '21 - PROCEEDINGS OF THE 44TH INTERNATIONAL ACM SIGIR CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION RETRIEVAL | 2021年
基金
澳大利亚研究理事会;
关键词
Directional Graph Convolutional Networks; Structural Information; Random Walk with Restart;
D O I
10.1145/3404835.3462834
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Graph structural information such as topologies or connectivities provides valuable guidance for graph convolutional networks (GCNs) to learn nodes' representations. Existing GCN models that capture nodes' structural information weight in- and out-neighbors equally or differentiate in- and out-neighbors globally without considering nodes' local topologies. We observe that in- and out-neighbors contribute differently for nodes with different local topologies. To explore the directional structural information for different nodes, we propose a GCN model with weighted structural features, named WGCN. WGCN first captures nodes' structural fingerprints via a direction and degree aware Random Walk with Restart algorithm, where the walk is guided by both edge direction and nodes' in- and out-degrees. Then, the interactions between nodes' structural fingerprints are used as the weighted node structural features. To further capture nodes' high-order dependencies and graph geometry, WGCN embeds graphs into a latent space to obtain nodes' latent neighbors and geometrical relationships. Based on nodes' geometrical relationships in the latent space, WGCN differentiates latent, in-, and out-neighbors with an attention-based geometrical aggregation. Experiments on transductive node classification tasks show that WGCN outperforms the baseline models consistently by up to 17.07% in terms of accuracy on five benchmark datasets.
引用
收藏
页码:624 / 633
页数:10
相关论文
共 54 条
[41]  
Tong Zekun, 2020, ADV NEUR IN, V33
[42]  
Velickovic P, 2018, P INT C LEARNING REP
[43]   Heterogeneous Graph Attention Network [J].
Wang, Xiao ;
Ji, Houye ;
Shi, Chuan ;
Wang, Bai ;
Cui, Peng ;
Yu, P. ;
Ye, Yanfang .
WEB CONFERENCE 2019: PROCEEDINGS OF THE WORLD WIDE WEB CONFERENCE (WWW 2019), 2019, :2022-2032
[44]   NodeAug: Semi-Supervised Node Classification with Data Augmentation [J].
Wang, Yiwei ;
Wang, Wei ;
Liang, Yuxuan ;
Cai, Yujun ;
Liu, Juncheng ;
Hooi, Bryan .
KDD '20: PROCEEDINGS OF THE 26TH ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY & DATA MINING, 2020, :207-217
[45]   Graph-based Semi-supervised Learning for Text Classification [J].
Widmann, Natalie ;
Verberne, Suzan .
ICTIR'17: PROCEEDINGS OF THE 2017 ACM SIGIR INTERNATIONAL CONFERENCE THEORY OF INFORMATION RETRIEVAL, 2017, :59-66
[46]   Incorporating Graph Attention and Recurrent Architectures for City-Wide Taxi Demand Prediction [J].
Xu, Ying ;
Li, Dongsheng .
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION, 2019, 8 (09)
[47]  
Zhang Kai, 2020, ICLR
[48]   HexCNN: A Framework for Native Hexagonal Convolutional Neural Networks [J].
Zhao, Yunxiang ;
Ke, Qiuhong ;
Korn, Flip ;
Qi, Jianzhong ;
Zhang, Rui .
20TH IEEE INTERNATIONAL CONFERENCE ON DATA MINING (ICDM 2020), 2020, :1424-1429
[49]   CBHE: Corner-based Building Height Estimation for Complex Street Scene Images [J].
Zhao, Yunxiang ;
Qi, Jianzhong ;
Zhang, Rui .
WEB CONFERENCE 2019: PROCEEDINGS OF THE WORLD WIDE WEB CONFERENCE (WWW 2019), 2019, :2436-2447
[50]  
Zhibin Lu, 2020, Advances in Information Retrieval, 42nd European Conference on IR Research, ECIR 2020. Proceedings. Lecture Notes in Computer Science (LNCS 12035), P369, DOI 10.1007/978-3-030-45439-5_25