Multi-scale attributed node embedding

Benedek Rozemberczki*, Carl Allen, Rik Sarkar

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


We present network embedding algorithms that capture information about a node from the local distribution over node attributes around it, as observed over random walks following an approach similar to Skip-gram. Observations from neighbourhoods of different sizes are either pooled (AE) or encoded distinctly in a multi-scale approach (MUSAE). Capturing attribute-neighbourhood relationships over multiple scales is useful for a range of applications, including latent feature identification across disconnected networks with similar features. We prove theoretically that matrices of node-feature pointwise mutual information are implicitly factorized by the embeddings. Experiments show that our algorithms are computationally efficient and outperform comparable models on social networks and web graphs.

Original languageEnglish
Pages (from-to)1-22
Number of pages22
JournalJournal of Complex Networks
Issue number2
Publication statusPublished - 7 May 2021


  • attributed network
  • dimensionality reduction
  • node classification
  • node embedding


Dive into the research topics of 'Multi-scale attributed node embedding'. Together they form a unique fingerprint.

Cite this