Skip to main navigation Skip to search Skip to main content

BDNE: A method of bi-directional distance network embedding

  • Dongjie Zhu
  • , Yundong Sun
  • , Ning Cao
  • , Xueming Qiao
  • , Ming Xu
  • , Jinlin Li
  • , Junzhou Yang
  • Harbin Institute of Technology Weihai
  • State Grid Shandong Electric Power Company

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In order to capture the directed relationship between nodes more accurately, this paper proposed a novel network embedding model called BDNE. The model adds the bi-directional distance while preserving the co-occurrence frequency of the nodes within a context window, which is of great significance in some applications of social networks, such as public opinion monitoring and control, and group discovery. Experimental results show that the embedding results of our model as features in the node classification task are better than DeepWalk, Node2Vector and Line on real data sets of different types and sizes.

Original languageEnglish
Title of host publicationProceedings - 2019 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery, CyberC 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages158-161
Number of pages4
ISBN (Electronic)9781728125411
DOIs
StatePublished - Oct 2019
Externally publishedYes
Event2019 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery, CyberC 2019 - Guilin, China
Duration: 17 Oct 201919 Oct 2019

Publication series

NameProceedings - 2019 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery, CyberC 2019

Conference

Conference2019 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery, CyberC 2019
Country/TerritoryChina
CityGuilin
Period17/10/1919/10/19

Keywords

  • Data mining
  • Network embedding
  • Representation learning

Fingerprint

Dive into the research topics of 'BDNE: A method of bi-directional distance network embedding'. Together they form a unique fingerprint.

Cite this