Unifying community detection and network embedding in attributed networks

• Published in: Knowledge and information systems Vol. 63; no. 5; pp. 1221 - 1239
• Main Authors: Ding, Yu, Wei, Hao, Hu, Guyu, Pan, Zhisong, Wang, Shuaihui
• Format: Journal Article
• Language: English
• Published: London Springer London 01.05.2021; Springer Nature B.V
• Subjects: Computer Science; Data Mining and Knowledge Discovery; Database Management; Embedding; Information Storage and Retrieval; Information Systems and Communication Service; Information Systems Applications (incl.Internet); IT in Business; Nodes; Regular Paper; Attributed network; Feature learning; Network embedding; Attributed network embedding; Network representation
• ISSN: 0219-1377; 0219-3116
• DOI: 10.1007/s10115-021-01557-5

Traditionally, community detection and network embedding are two separate tasks. Network embedding aims to output a vector representation for each node in the network, and community detection aims to find all densely connected groups of nodes and well separate them from others. Most of the existing approaches do community detection and network embedding in a separate manner, and ignore node attributes information, which leads to poor results. In this paper, we propose a novel model that jointly solves the network embedding and community detection problems together. The model can make use of the network local information, the global information and node attributes information collaboratively. We empirically show that by jointly solving these two problems together, the model can greatly improve the ability of community detection, but also learn better network embedding than the advanced baseline methods. We evaluate the proposed model on several datasets, and the experimental results have shown the effectiveness and advancement of our model.