SENA: Preserving social structure for network embedding

Sanghyun Hong, Tanmoy Chakraborty, Sungjin Ahn, Ghaith Husari, Noseong Park

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

Network embedding transforms a network into a continuous feature space where inherent properties of the network are preserved. Network augmentation, on the other hand, leverages this feature representation to obtain a more informative network by adding potentially plausible edges while removing noisy edges. Traditional network embedding methods are often inefficient in capturing - (i) the latent relationship when the network is sparse (the network sparsity problem), and (ii) the local and global neighborhood structure of vertices unique to the network (structure preserving problem). In this paper, we propose SENA, a structural embedding and network augmentation framework for social network analysis. Unlike existing social embedding methods which only generate vertex features, SENA generates features for both vertices and relations (edges) after solving the aforementioned two problems. We compare SENA with four baseline network embedding methods, namely DeepWalk, SE, SME and TransE. We demonstrate the efficacy of SENA through a task-based evaluation setting on different real-world networks. We achieve up to 13.67% higher accuracy for community detection and link prediction.

Original languageEnglish
Title of host publicationHT 2017 - Proceedings of the 28th ACM Conference on Hypertext and Social Media
PublisherAssociation for Computing Machinery, Inc
Pages235-244
Number of pages10
ISBN (Electronic)9781450347082
DOIs
Publication statusPublished - 2017 Jul 4
Event28th ACM Conference on Hypertext and Social Media, HT 2017 - Prague, Czech Republic
Duration: 2017 Jul 42017 Jul 7

Publication series

NameHT 2017 - Proceedings of the 28th ACM Conference on Hypertext and Social Media

Conference

Conference28th ACM Conference on Hypertext and Social Media, HT 2017
CountryCzech Republic
CityPrague
Period17/7/417/7/7

All Science Journal Classification (ASJC) codes

  • Computer Graphics and Computer-Aided Design
  • Artificial Intelligence
  • Human-Computer Interaction
  • Software

Fingerprint Dive into the research topics of 'SENA: Preserving social structure for network embedding'. Together they form a unique fingerprint.

  • Cite this

    Hong, S., Chakraborty, T., Ahn, S., Husari, G., & Park, N. (2017). SENA: Preserving social structure for network embedding. In HT 2017 - Proceedings of the 28th ACM Conference on Hypertext and Social Media (pp. 235-244). (HT 2017 - Proceedings of the 28th ACM Conference on Hypertext and Social Media). Association for Computing Machinery, Inc. https://doi.org/10.1145/3078714.3078738