Skip to content
/ WFD Public

Weighted flow diffusion for local graph clustering with node attributes

Notifications You must be signed in to change notification settings

s-h-yang/WFD

Repository files navigation

Weighted Flow Diffusion

Code to reproduce the empirical results in our paper Weighted Flow Diffusion for Local Graph Clustering with Node Attributes: an Algorithm and Statistical Guarantees.

  • The notebook synthetic.ipynb reproduces Figure 1 and Figure 2 in the paper.
  • To reproduce Tables 1, 3, 5 in the paper, run Julia script experiment-{dataset}.jl. This will produce two output text files. F1-{datasset}-base.txt stores average F1 scores without using node attributes. F1-{dataset}-wfd.txt stores average F1 scores using node attributes.

About

Weighted flow diffusion for local graph clustering with node attributes

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published