TransN: Heterogeneous Network Representation Learning by Translating Node Embeddings

被引:10
作者
Li, Zijian [1 ]
Zheng, Wenhao [2 ]
Lin, Xueling [1 ]
Zhao, Ziyuan [3 ]
Wang, Zhe [1 ]
Wang, Yue [4 ]
Jian, Xun [1 ]
Chen, Lei [1 ]
Yan, Qiang [3 ]
Mao, Tiezheng [3 ]
机构
[1] HKUST, Comp Sci & Engn Dept, Hong Kong, Peoples R China
[2] Nanjing Univ, Natl Key Lab Novel Software Technol, Nanjing, Peoples R China
[3] Tencent Corp, Wechat Grp, Guangzhou, Peoples R China
[4] Shenzhen Univ, Shenzhen Inst Comp Sci, Shenzhen, Peoples R China
来源
2020 IEEE 36TH INTERNATIONAL CONFERENCE ON DATA ENGINEERING (ICDE 2020) | 2020年
基金
美国国家科学基金会;
关键词
Heterogeneous Network Embedding; Representation Learning; Multi-View Network Embedding; Dual Learning;
D O I
10.1109/ICDE48307.2020.00057
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Learning network embeddings has attracted growing attention in recent years. However, most of the existing methods focus on homogeneous networks, which cannot capture the important type information in heterogeneous networks. To address this problem, in this paper, we propose TransN, a novel multi-view network embedding framework for heterogeneous networks. Compared with the existing methods, TransN is an unsupervised framework which does not require node labels or user-specified meta-paths as inputs. In addition, TransN is capable of handling more general types of heterogeneous networks than the previous works. Specifically, in our framework TransN, we propose a novel algorithm to capture the proximity information inside each single view. Moreover, to transfer the learned information across views, we propose an algorithm to translate the node embeddings between different views based on the dual-learning mechanism, which can both capture the complex relations between node embeddings in different views, and preserve the proximity information inside each view during the translation. We conduct extensive experiments on real-world heterogeneous networks, whose results demonstrate that the node embeddings generated by TransN outperform those of competitors in various network mining tasks.
引用
收藏
页码:589 / 600
页数:12
相关论文
共 51 条
  • [1] [Anonymous], 2017, ARXIV PREPRINT ARXIV
  • [2] [Anonymous], 2016, Deep learning. vol
  • [3] [Anonymous], 2018, ARXIV180410850
  • [4] [Anonymous], 2019, ARXIV180308375V2 CSN
  • [5] CORRELATED RANDOM-WALKS
    BENDER, EA
    RICHMOND, LB
    [J]. ANNALS OF PROBABILITY, 1984, 12 (01) : 274 - 278
  • [6] Bishop C., 2006, PATTERN RECOGN, DOI DOI 10.1117/1.2819119
  • [7] Bordes Antoine, 2013, ADV NEURAL INF PROCE, P2787
  • [8] 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
  • [9] Cao SS, 2016, AAAI CONF ARTIF INTE, P1145
  • [10] Heterogeneous Network Embedding via Deep Architectures
    Chang, Shiyu
    Han, Wei
    Tang, Jiliang
    Qi, Guo-Jun
    Aggarwal, Charu C.
    Huang, Thomas S.
    [J]. KDD'15: PROCEEDINGS OF THE 21ST ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, 2015, : 119 - 128