Class NMFConsensusMatrix uses a non-negative matrix factorization to cluster samples.

Namespace:  CenterSpace.NMath.Stats
Assembly:  NMathStats (in NMathStats.dll) Version: 3.4.0.0

Syntax

C#
public class NMFConsensusMatrix<Alg> : ConnectivityMatrix
where Alg : new(), INMFUpdateAlgorithm
Visual Basic (Declaration)
Public Class NMFConsensusMatrix(Of Alg As {New, INMFUpdateAlgorithm}) _
	Inherits ConnectivityMatrix
Visual C++
generic<typename Alg>
where Alg : gcnew(), INMFUpdateAlgorithm
public ref class NMFConsensusMatrix : public ConnectivityMatrix

Type Parameters

Alg
Update algorithm for the non-negative matrix factorization.

Remarks

Since the non-negative matrix factorization may not converge to the same solution each time it is run, several runs are made and the average is computed.

For each run of the clustering non-negative matrix factorization a connectivity matrix C with cij = 1 if samples i and j belong to the same cluster and cij = 0 if they belong to different clusters. The consensus matrix is then computed as the average connectivity matrix over several clustering runs. The values of the consensus matrix range in value from 0 to 1 and reflect the probability that samples i and j cluster together.

Inheritance Hierarchy

System..::.Object
  CenterSpace.NMath.Matrix..::.DoubleSymmetricMatrix
    CenterSpace.NMath.Stats..::.ConnectivityMatrix
      CenterSpace.NMath.Stats..::.NMFConsensusMatrix<(Of <(Alg>)>)

See Also