-
Notifications
You must be signed in to change notification settings - Fork 4
Walktrap (WT by Pons and Latapy)
Sergiu Tripon edited this page Jul 27, 2016
·
5 revisions
Home ▸ Algorithms ▸ Walktrap (WT by Pons and Latapy)
Walktrap uses a hierarchical agglomerative method like FastGreedy (FG) but with a different merging criterion. Unlike FG, which relies on the modularity measure, WT uses a node-to-node distance measure to identify the closest communities.
This distance is based on the concept of random-walk. If two nodes are in the same community, the probability to get to a third one located in the same community through a random walk should not be very different for both of them. The distance is constructed by summing these differences over all nodes, with a correction for degree.
Home |
Scope |
Research |
Value |
Background
Literature Survey | Data | Algorithms | Plan | Meetings Journal
Literature Survey | Data | Algorithms | Plan | Meetings Journal
© Copyright 2016 Sergiu Tripon. All Rights Reserved.