L1-regularized PPR on directed graphs?

[bkj]

Hello --

I'm interested in trying to apply the L1-regularized PPR on a directed graph ... do you know if there's an implementation of that somewhere? Or - if not - is there a modification to the undirected algorithm that I could try to implement? I tried playing around w/ the math a bit myself ... but didn't make much progress.

Thanks!

[dgleich]

Just drop direction is the easiest...

If you want to prefer something with direction you need to articulate what that is… see local higher order graph clustering for some examples

Easy way: just pick the most meaningful directed triangle and use that as your motif.

This is short for a much longer discussion… e-mail if you want to chat about it.

On Wed, Aug 3, 2022 at 10:18 Ben Johnson @.***> wrote:

Hello --

I'm interested in trying to apply the L1-regularized PPR on a directed graph ... do you know if there's an implementation of that somewhere? Or - if not - is there a modification to the undirected algorithm that I could try to implement? I tried playing around w/ the math a bit myself ... but didn't make much progress.

Thanks!

— Reply to this email directly, view it on GitHub https://github.com/kfoynt/LocalGraphClustering/issues/127, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAIJS7IMEQ6RK4XTHXQU2DVXKLVJANCNFSM55POLM4A . You are receiving this because you are subscribed to this thread.Message ID: @.***>

[kfoynt]

I am adding a bit of info to that too.

It’s easy to use higher order method for undirected graphs. Shenghao (CC’ed) has an implementation here: https://github.com/opallab/Local-Hyper-flow-Diffusion You can follow up with him on how to use it if you want.

On Aug 3, 2022, at 12:45 PM, David Gleich @.@.>> wrote:

Just drop direction is the easiest...

If you want to prefer something with direction you need to articulate what that is… see local higher order graph clustering for some examples

Easy way: just pick the most meaningful directed triangle and use that as your motif.

This is short for a much longer discussion… e-mail if you want to chat about it.

On Wed, Aug 3, 2022 at 10:18 Ben Johnson @.***> wrote:

Hello --

I'm interested in trying to apply the L1-regularized PPR on a directed graph ... do you know if there's an implementation of that somewhere? Or - if not - is there a modification to the undirected algorithm that I could try to implement? I tried playing around w/ the math a bit myself ... but didn't make much progress.

Thanks!

— Reply to this email directly, view it on GitHub https://github.com/kfoynt/LocalGraphClustering/issues/127, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAIJS7IMEQ6RK4XTHXQU2DVXKLVJANCNFSM55POLM4A . You are receiving this because you are subscribed to this thread.Message ID: @.***>

— Reply to this email directly, view it on GitHubhttps://github.com/kfoynt/LocalGraphClustering/issues/127#issuecomment-1204215505, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ACZN7WIE2MO7HSLAJYHYZQTVXKO2DANCNFSM55POLM4A. You are receiving this because you are subscribed to this thread.Message ID: @.***>