Class NormalizedCutSpectralClustering<U>
java.lang.Object
es.uam.eps.ir.relison.sna.community.detection.modularity.balanced.SpectralClustering<U>
es.uam.eps.ir.relison.sna.community.detection.modularity.balanced.NormalizedCutSpectralClustering<U>
- Type Parameters:
U
- Type of the users.
- All Implemented Interfaces:
CommunityDetectionAlgorithm<U>
public class NormalizedCutSpectralClustering<U> extends SpectralClustering<U>
Community detection algorithm for balanced communities. It uses the normalized cut Laplacian.
Reference: R. Zafarani, M.A. Abassi, H. Liu. Social Media Mining: An Introduction. Chapter 6. 2014
-
Constructor Summary
Constructors Constructor Description NormalizedCutSpectralClustering(int k)
Constructor. -
Method Summary
Modifier and Type Method Description protected cern.colt.matrix.DoubleMatrix2D
laplacian(cern.colt.matrix.DoubleMatrix2D degree, cern.colt.matrix.DoubleMatrix2D adjacency)
Computes the Laplacian matrix of a graph, given the degree and adjacency matrices.Methods inherited from class es.uam.eps.ir.relison.sna.community.detection.modularity.balanced.SpectralClustering
detectCommunities
-
Constructor Details
-
NormalizedCutSpectralClustering
public NormalizedCutSpectralClustering(int k)Constructor.- Parameters:
k
- The number of clusters we want to find
-
-
Method Details
-
laplacian
protected cern.colt.matrix.DoubleMatrix2D laplacian(cern.colt.matrix.DoubleMatrix2D degree, cern.colt.matrix.DoubleMatrix2D adjacency)Description copied from class:SpectralClustering
Computes the Laplacian matrix of a graph, given the degree and adjacency matrices.- Specified by:
laplacian
in classSpectralClustering<U>
- Parameters:
degree
- the degree matrix of a graphadjacency
- the adjacency matrix of a graph.- Returns:
- the Laplacian matrix.
-