XMeansOptions<T>
Options & Configuration · AiDotNet.Clustering.Options
Configuration options for X-Means clustering.
Inheritance: ModelOptions → ClusteringOptions<T> → XMeansOptions<T>
For Beginners
X-Means finds K automatically.
The problem with K-Means: You must choose K (number of clusters). X-Means solves this by:
- Start with a small K
- Try splitting each cluster into two
- Keep the split if it improves the model (using BIC)
- Stop when no splits improve the model
BIC (Bayesian Information Criterion) balances:
- Model fit (how well clusters explain data)
- Model complexity (penalizes too many clusters)
How It Works
X-Means extends K-Means by automatically determining the optimal number of clusters using the Bayesian Information Criterion (BIC). It starts with a minimum number of clusters and iteratively splits clusters until BIC stops improving.
Constructors
| Constructor | Summary |
|---|---|
XMeansOptions() |
Initializes XMeansOptions with appropriate defaults. |
Properties
| Property | Summary |
|---|---|
InformationCriterion Criterion { get; set; } |
Gets or sets the information criterion to use. |
int MaxClusters { get; set; } |
Gets or sets the maximum number of clusters. |
int MinClusters { get; set; } |
Gets or sets the minimum number of clusters. |