@saehrimnir/druidjs / CURE

Class: CURE

Defined in: clustering/CURE.js:17

CURE (Clustering Using REpresentatives)

An efficient clustering algorithm for large databases that is robust to outliers and identifies clusters with non-spherical shapes and wide variances in size.

Extends

• Clustering

Constructors

Constructor

new CURE(points: InputType, parameters?: Partial<ParametersCURE>): CURE;

Defined in: clustering/CURE.js:36

Parameters

Parameter	Type	Description
points	InputType	-
parameters	Partial<ParametersCURE>	-

Returns

CURE

Overrides

Clustering.constructor

Properties

Property	Type	Default value	Inherited from	Defined in
_cluster_ids	number[]	[]	-	clustering/CURE.js:30
_D	number	undefined	Clustering._D	clustering/Clustering.js:19
_K	number	undefined	-	clustering/CURE.js:19
_matrix	Matrix	undefined	Clustering._matrix	clustering/Clustering.js:15
_N	number	undefined	Clustering._N	clustering/Clustering.js:17
_num_representatives	number	undefined	-	clustering/CURE.js:21
_parameters	ParametersCURE	undefined	Clustering._parameters	clustering/Clustering.js:13
_points	InputType	undefined	Clustering._points	clustering/Clustering.js:11
_shrink_factor	number	undefined	-	clustering/CURE.js:23

Methods

get_cluster_list()

get_cluster_list(): number[];

Defined in: clustering/CURE.js:250

Returns

number[]

Overrides

Clustering.get_cluster_list

---

get_clusters()

get_clusters(): number[][];

Defined in: clustering/CURE.js:243

Returns

number[][]

Overrides

Clustering.get_clusters