org.apache.mahout.clustering.iterator
Interface ClusteringPolicy

All Superinterfaces:
org.apache.hadoop.io.Writable
All Known Implementing Classes:
AbstractClusteringPolicy, CanopyClusteringPolicy, FuzzyKMeansClusteringPolicy, KMeansClusteringPolicy

public interface ClusteringPolicy
extends org.apache.hadoop.io.Writable

A ClusteringPolicy captures the semantics of assignment of points to clusters


Method Summary
 Vector classify(Vector data, ClusterClassifier prior)
          Classify the data vector given the classifier's models
 void close(ClusterClassifier posterior)
          Close the policy using the classifier's models
 Vector select(Vector probabilities)
          Return a vector of weights for each of the models given those probabilities
 void update(ClusterClassifier posterior)
          Update the policy with the given classifier
 
Methods inherited from interface org.apache.hadoop.io.Writable
readFields, write
 

Method Detail

classify

Vector classify(Vector data,
                ClusterClassifier prior)
Classify the data vector given the classifier's models

Parameters:
data - a data Vector
prior - a prior ClusterClassifier
Returns:
a Vector of probabilities that the data is described by each of the models

select

Vector select(Vector probabilities)
Return a vector of weights for each of the models given those probabilities

Parameters:
probabilities - a Vector of pdfs
Returns:
a Vector of weights

update

void update(ClusterClassifier posterior)
Update the policy with the given classifier

Parameters:
posterior - a ClusterClassifier

close

void close(ClusterClassifier posterior)
Close the policy using the classifier's models

Parameters:
posterior - a posterior ClusterClassifier


Copyright © 2008–2014 The Apache Software Foundation. All rights reserved.