Clusteval logo ClustEval clustering evaluation framework

Here you see general information about the clustering method.

Clustering Method 'Markov Clustering' - General

Markov Clustering

Markov In Markov Clustering the input similarities are interpret as a graph on which densely connected areas are identified by simulating random walks. The clustering processes itself is performed by repeatedly alternating so-called expansion steps and inflation steps. The expansion steps allow simulated walks of long distances while the inflation steps sharpen the emerging local cluster structure.

  • Publication: Stijn Dongen. A cluster algorithm for graphs. Technical report, Amsterdam, The Netherlands, The Netherlands, 2000.

Information Value