ConsensusClusteringOptions<T>

Options & Configuration ยท AiDotNet.Clustering.Ensemble

Configuration options for Consensus (Ensemble) Clustering.

Inheritance: ModelOptions โ†’ ConsensusClusteringOptions<T>

For Beginners

Ensemble clustering is like taking a vote.

The idea:

  1. Run multiple clustering algorithms (or same algorithm multiple times)
  2. Each gives a different answer
  3. Combine the answers to get a more reliable result

Why it works:

  • Different algorithms have different strengths
  • Random initialization can give different results
  • Combining reduces the impact of any single bad result

Common approaches:

  • Co-association matrix: How often are points clustered together?
  • Voting: Which cluster assignment is most popular?

How It Works

Consensus clustering combines multiple clustering solutions to produce a more robust final clustering. It works by aggregating partitions from different algorithms or the same algorithm with different parameters.

Constructors

Constructor Summary
ConsensusClusteringOptions()

Properties

Property Summary
ConsensusMethod Method { get; set; } Gets or sets the consensus method.
FinalClusteringAlgorithm FinalAlgorithm { get; set; } Gets or sets the final clustering algorithm.
int NumBaseClusterings { get; set; } Gets or sets the number of base clusterings to generate.
int? NumClusters { get; set; } Gets or sets the target number of clusters.