Package gov.sandia.cognition.learning.algorithm.clustering.initializer

Provides implementations of methods for selecting initial clusters.

See:
          Description

Interface Summary
FixedClusterInitializer<ClusterType extends Cluster<DataType>,DataType> The FixedClusterInitializer interface defines the functionality of a class that can initialize a given number of clusters from a set of elements.
 

Class Summary
AbstractMinDistanceFixedClusterInitializer<ClusterType extends Cluster<DataType>,DataType> Implements an abstract FixedClusterInitializer that works by using the minimum distance from a point to the cluster.
DistanceSamplingClusterInitializer<ClusterType extends Cluster<DataType>,DataType> Implements FixedClusterInitializer that initializes clusters by first selecting a random point for the first cluster and then randomly sampling each successive cluster based on the squared minimum distance from the point to the existing selected clusters.
GreedyClusterInitializer<ClusterType extends Cluster<DataType>,DataType> Implements a FixedClusterInitializer that greedily attempts to create the initial clusters.
NeighborhoodGaussianClusterInitializer Creates GaussianClusters near existing, but not on top of, data points.
 

Package gov.sandia.cognition.learning.algorithm.clustering.initializer Description

Provides implementations of methods for selecting initial clusters. These provide some clustering algorithms, such as k-means with a starting set of clusters based on the data.

Since:
2.0
Author:
Justin Basilico