Click or drag to resize

WardsMinimumVarianceLinkageTInstance Class

Implements Ward's minimum variance method, i.e., returns the total within-cluster variance, corresponding to a weighted squared distance between cluster centers.
Inheritance Hierarchy
SystemObject
  Aglomera.LinkageWardsMinimumVarianceLinkageTInstance

Namespace:  Aglomera.Linkage
Assembly:  Aglomera (in Aglomera.dll) Version: 1.1.0
Syntax
C#
Copy
public class WardsMinimumVarianceLinkage<TInstance> : ILinkageCriterion<TInstance>
where TInstance : Object, IComparable<TInstance>
Request Example View Source

Type Parameters

TInstance
The type of instance considered.

The WardsMinimumVarianceLinkageTInstance type exposes the following members.

Constructors
  NameDescription
Public methodWardsMinimumVarianceLinkageTInstance
Creates a new WardsMinimumVarianceLinkageTInstance with given dissimilarity metric and centroid function.
Top
Properties
  NameDescription
Public propertyDissimilarityMetric
Gets the metric used to measure the dissimilarity / distance between cluster elements according to this linkage criterion.
Top
Methods
  NameDescription
Public methodCalculate
Calculates the distance / dissimilarity between the two given clusters according to this linkage criterion.
Public methodEquals (Inherited from Object.)
Protected methodFinalize (Inherited from Object.)
Public methodGetHashCode (Inherited from Object.)
Public methodGetType (Inherited from Object.)
Protected methodMemberwiseClone (Inherited from Object.)
Public methodToString (Inherited from Object.)
Top
Remarks
"With hierarchical clustering, the sum of squares starts out at zero (because every point is in its own cluster) and then grows as we merge clusters. Ward's method keeps this growth as small as possible. This is nice if you believe that the sum of squares should be small. Notice that the number of points shows up in [the formula], as well as their geometric separation. Given two pairs of clusters whose centers are equally far apart, Ward's method will prefer to merge the smaller ones." [1] References: [1] - http://www.stat.cmu.edu/~cshalizi/350/lectures/08/lecture-08.pdf.
See Also