/MarkovClustering

Markov clustering algorithm

Primary LanguageJava

Markov Cluster Algorithm (MCL)

Author: Jiri Krizek

Supervisor: Jaroslav Kuchar

The Markov Clustering plugin for Gephi. This plugin finds clusters in graph, which can be used in Social Network Analysis.

Clustering on Graphs: The Markov Cluster Algorithm (MCL)

Tutorial

You can start cluster finding using "Clustering" panel. This panel is usually on the left part of Gephi window. If you don't see this panel, enable it using "Window/Clustering" from the main menu.

From dropdown menu, select Markov clustering.

Then you can edit algorithm parameters using button Settings: mclparams

Parameters

  • Add self loops to each node - improves clusters computation (small simple path loops can complicate computation in odd powers, enabling this parameter resolves this)
  • Generate nodes with more clusters as standalone cluster - In special cases (symmetric graph), node can fall into two different clusters at the same time. You can use this parameter to show this multiple nodes as new cluster (e. g. Clusters: [1, 2] - 1 node)
  • Power parameter - affects time of computation and partially cluster granularity
  • Inflation parameter - affects cluster granularity (bigger the number means more clusters)
  • Pruning parameter - computed values below this number will be interpreted as zero values, this parameter speeds up the computation

After setting the parameters, you can start computation using Run button in the Clustering panel.

License

The GPL version 3, http://www.gnu.org/licenses/gpl-3.0.txt