Graph Autoencoder with Preserving Node Attribute Similarity

被引:9
作者
Lin, Mugang [1 ,2 ]
Wen, Kunhui [1 ]
Zhu, Xuanying [1 ]
Zhao, Huihuang [1 ,2 ]
Sun, Xianfang [3 ]
机构
[1] Hengyang Normal Univ, Coll Comp Sci & Technol, Hengyang 421002, Peoples R China
[2] Hunan Prov Key Lab Intelligent Informat Proc & App, Hengyang 421002, Peoples R China
[3] Cardiff Univ, Sch Comp Sci & Informat, Cardiff CF24 4AG, Wales
基金
中国国家自然科学基金;
关键词
graph representation learning; graph autoencoder; unsupervised learning; k-nearest neighbor;
D O I
10.3390/e25040567
中图分类号
O4 [物理学];
学科分类号
0702 ;
摘要
The graph autoencoder (GAE) is a powerful graph representation learning tool in an unsupervised learning manner for graph data. However, most existing GAE-based methods typically focus on preserving the graph topological structure by reconstructing the adjacency matrix while ignoring the preservation of the attribute information of nodes. Thus, the node attributes cannot be fully learned and the ability of the GAE to learn higher-quality representations is weakened. To address the issue, this paper proposes a novel GAE model that preserves node attribute similarity. The structural graph and the attribute neighbor graph, which is constructed based on the attribute similarity between nodes, are integrated as the encoder input using an effective fusion strategy. In the encoder, the attributes of the nodes can be aggregated both in their structural neighborhood and by their attribute similarity in their attribute neighborhood. This allows performing the fusion of the structural and node attribute information in the node representation by sharing the same encoder. In the decoder module, the adjacency matrix and the attribute similarity matrix of the nodes are reconstructed using dual decoders. The cross-entropy loss of the reconstructed adjacency matrix and the mean-squared error loss of the reconstructed node attribute similarity matrix are used to update the model parameters and ensure that the node representation preserves the original structural and node attribute similarity information. Extensive experiments on three citation networks show that the proposed method outperforms state-of-the-art algorithms in link prediction and node clustering tasks.
引用
收藏
页数:22
相关论文
共 45 条
  • [1] Graph Regularized Autoencoder and its Application in Unsupervised Anomaly Detection
    Ahmed, Imtiaz
    Galoppo, Travis
    Hu, Xia
    Ding, Yu
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2022, 44 (08) : 4110 - 4124
  • [2] A Comprehensive Survey of Graph Embedding: Problems, Techniques, and Applications
    Cai, HongYun
    Zheng, Vincent W.
    Chang, Kevin Chen-Chuan
    [J]. IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2018, 30 (09) : 1616 - 1637
  • [3] Cao S., 2015, PROC 24 INTERNAT C I, P891, DOI DOI 10.1145/2806416.2806512
  • [4] metapath2vec: Scalable Representation Learning for Heterogeneous Networks
    Dong, Yuxiao
    Chawla, Nitesh V.
    Swami, Ananthram
    [J]. KDD'17: PROCEEDINGS OF THE 23RD ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, 2017, : 135 - 144
  • [5] Graph autoencoder-based unsupervised outlier detection
    Du, Xusheng
    Yu, Jiong
    Chu, Zheng
    Jin, Lina
    Chen, Jiaying
    [J]. INFORMATION SCIENCES, 2022, 608 : 532 - 550
  • [6] Dwivedi V.P., 2022, P 10 INT C LEARNING
  • [7] Gao HC, 2018, PROCEEDINGS OF THE TWENTY-SEVENTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, P3364
  • [8] Grover A, 2019, PR MACH LEARN RES, V97
  • [9] node2vec: Scalable Feature Learning for Networks
    Grover, Aditya
    Leskovec, Jure
    [J]. KDD'16: PROCEEDINGS OF THE 22ND ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, 2016, : 855 - 864
  • [10] Multi-Scale Variational Graph AutoEncoder for Link Prediction
    Guo, Zhihao
    Wang, Feng
    Yao, Kaixuan
    Liang, Jiye
    Wang, Zhiqiang
    [J]. WSDM'22: PROCEEDINGS OF THE FIFTEENTH ACM INTERNATIONAL CONFERENCE ON WEB SEARCH AND DATA MINING, 2022, : 334 - 342