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:
- Run multiple clustering algorithms (or same algorithm multiple times)
- Each gives a different answer
- 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. |