Click or drag to resize

SilhouetteCoefficientTInstance Class

Implements an internal evaluation method that measures how similar an element is to its own cluster (cohesion) compared to other clusters (separation). The silhouette ranges from −1 to +1, where a high value indicates that the element is well matched to its own cluster and poorly matched to neighboring clusters. If most elements (average) have a high value, then the clustering configuration is appropriate. If the average is a low or negative value, then the clustering configuration may have too many or too few clusters.
Inheritance Hierarchy
SystemObject
  Aglomera.Evaluation.InternalSilhouetteCoefficientTInstance

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

Type Parameters

TInstance
The type of instance considered.

The SilhouetteCoefficientTInstance type exposes the following members.

Constructors
  NameDescription
Public methodSilhouetteCoefficientTInstance
Creates a new SilhouetteCoefficientTInstance with given dissimilarity metric.
Top
Properties
  NameDescription
Public propertyDissimilarityMetric
Gets the metric used by this criterion to measure the dissimilarity / distance between cluster elements.
Top
Methods
Remarks
Assumptions in [2]: - "Note that the construction [...] depends on the availability of other clusters apart from A, so we have to assume [...] that the number of clusters k is more than one." - "When cluster A contains only a single object it is unclear how u(i) should be defined, and then we simply set s(i) equal to zero. This choice is of course arbitrary, but a value of zero appears to be most neutral." Therefore, Silhouette coefficient punishes outliers and noise, so in the presence of such data we should avoid it. References: [1] - https://en.wikipedia.org/wiki/Silhouette_(clustering) [2] - Rousseeuw, P. J. (1987). Silhouettes: a graphical aid to the interpretation and validation of cluster analysis. Journal of computational and applied mathematics, 20, 53-65.
See Also