Network Embedding Using Deep Robust Nonnegative Matrix Factorization

被引:10
作者
He, Chaobo [1 ]
Liu, Hai [2 ]
Tang, Yong [2 ]
Fei, Xiang [3 ]
Li, Hanchao [3 ]
Zhang, Qiong [4 ]
机构
[1] Zhongkai Univ Agr & Engn, Sch Informat Sci & Technol, Guangzhou 510225, Peoples R China
[2] South China Normal Univ, Sch Comp Sci, Guangzhou 510631, Peoples R China
[3] Coventry Univ, Dept Comp, Coventry CV1 5FB, W Midlands, England
[4] Nanchang Inst Technol, Coll Comp Informat & Engn, Nanchang 330044, Jiangxi, Peoples R China
基金
中国国家自然科学基金;
关键词
Network embedding; deep nonnegative matrix factorization; network analysis; complex networks;
D O I
10.1109/ACCESS.2020.2992269
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
As an effective technique to learn low-dimensional node features in complicated network environment, network embedding has become a promising research direction in the field of network analysis. Due to the virtues of better interpretability and flexibility, matrix factorization based methods for network embedding have received increasing attentions. However, most of them are inadequate to learn more complicated hierarchical features hidden in complex networks because of their mechanisms of single-layer factorization structure. Besides, their original feature matrices used for factorization and their robustness against noises also need to be further improved. To solve these problems, we propose a novel network embedding method named DRNMF (deep robust nonnegative matrix factorization), which is formed by multi-layer NMF learning structure. Meanwhile, DRNMF employs the combination of high-order proximity matrices of the network as the original feature matrix for the factorization. To improve the robustness against noises, we use $\ell _{2,1}$ norm to devise the objective function for the DRNMF network embedding model. Effective iterative update rules are derived to resolve the model, and the convergence of these rules are strictly proved. Moreover, we introduce a pre-training strategy to improve the efficiency of convergence. Extensive experiments on several benchmarks of complex networks demonstrate that our proposed method DRNMF is effective and has better performance than the state-of-the-art matrix factorization based methods for network embedding.
引用
收藏
页码:85441 / 85453
页数:13
相关论文
共 39 条
  • [1] Aggarwal CC, 2011, SOCIAL NETWORK DATA ANALYTICS, P1, DOI 10.1007/978-1-4419-8462-3
  • [2] Ahmed A., 2013, WWW
  • [3] Boyd S., 2004, CONVEX OPTIMIZATION
  • [4] 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
  • [5] Cao S., 2015, P 24 ACM INT C INF K, P891, DOI DOI 10.1145/2806416.2806512
  • [6] Cao SS, 2016, AAAI CONF ARTIF INTE, P1145
  • [7] Convex and Semi-Nonnegative Matrix Factorizations
    Ding, Chris
    Li, Tao
    Jordan, Michael I.
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2010, 32 (01) : 45 - 55
  • [8] Community detection in networks: A user guide
    Fortunato, Santo
    Hric, Darko
    [J]. PHYSICS REPORTS-REVIEW SECTION OF PHYSICS LETTERS, 2016, 659 : 1 - 44
  • [9] Graph embedding techniques, applications, and performance: A survey
    Goyal, Palash
    Ferrara, Emilio
    [J]. KNOWLEDGE-BASED SYSTEMS, 2018, 151 : 78 - 94
  • [10] 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