@saehrimnir/druidjs / XMeans

Class: XMeans

Defined in: clustering/XMeans.js:34

X-Means Clustering

An extension of K-Means that automatically determines the number of clusters (K) using the Bayesian Information Criterion (BIC).

Extends

• Clustering

Constructors

Constructor

new XMeans(points: InputType, parameters?: Partial<ParametersXMeans>): XMeans;

Defined in: clustering/XMeans.js:54

XMeans clustering algorithm that automatically determines the optimal number of clusters.

X-Means extends K-Means by starting with a minimum number of clusters and iteratively splitting clusters to improve the Bayesian Information Criterion (BIC).

Algorithm:

1. Start with K_min clusters using KMeans
2. For each cluster, try splitting it into 2 sub-clusters
3. If BIC improves after splitting, keep the split
4. Run KMeans again with all (old + new) centroids
5. Repeat until K_max is reached or no more improvements

Parameters

Parameter	Type	Description
points	InputType	The data points to cluster
parameters?	Partial<ParametersXMeans>	Configuration parameters

Returns

XMeans

See

• https://www.cs.cmu.edu/~dpelleg/download/xmeans.pdf
• https://github.com/annoviko/pyclustering/blob/master/pyclustering/cluster/xmeans.py
• https://github.com/haifengl/smile/blob/master/core/src/main/java/smile/clustering/XMeans.java

Overrides

Clustering.constructor

Properties

Property	Type	Inherited from	Defined in
_best_kmeans	KMeans | null	-	clustering/XMeans.js:67
_D	number	Clustering._D	clustering/Clustering.js:19
_matrix	Matrix	Clustering._matrix	clustering/Clustering.js:15
_N	number	Clustering._N	clustering/Clustering.js:17
_parameters	ParametersXMeans	Clustering._parameters	clustering/Clustering.js:13
_points	InputType	Clustering._points	clustering/Clustering.js:11
_randomizer	Randomizer	-	clustering/XMeans.js:64

Accessors

centroids

Get Signature

get centroids(): Float64Array<ArrayBufferLike>[];

Defined in: clustering/XMeans.js:344

Get the final centroids

Returns

Float64Array<ArrayBufferLike>[]

Array of centroids

---

k

Get Signature

get k(): number;

Defined in: clustering/XMeans.js:356

Get the optimal number of clusters found

Returns

number

The number of clusters

Methods

get_cluster_list()

get_cluster_list(): number[];

Defined in: clustering/XMeans.js:332

Returns

number[]

The cluster list

Overrides

Clustering.get_cluster_list

---

get_clusters()

get_clusters(): number[][];

Defined in: clustering/XMeans.js:324

Get the computed clusters

Returns

number[][]

Array of clusters, each containing indices of points

Overrides

Clustering.get_clusters