gov.sandia.cognition.text.term.vector.weighter.normalize

Interface TermWeightNormalizer

All Superinterfaces:

java.lang.Cloneable, CloneableSerializable, java.io.Serializable

All Known Implementing Classes:

AbstractTermWeightNormalizer, UnitTermWeightNormalizer

public interface TermWeightNormalizer
extends CloneableSerializable

Interface for a tem weight normalization scheme.

Since:

3.0

Author:

Justin Basilico

Method Summary

Modifier and Type	Method and Description
void	normalizeWeights(Vector weights, Vector counts, Vector globalWeights) Normalizes the given weight vector.

Methods inherited from interface gov.sandia.cognition.util.CloneableSerializable

clone

Method Detail

normalizeWeights

void normalizeWeights(Vector weights,
                      Vector counts,
                      Vector globalWeights)

Normalizes the given weight vector. Can make use of the supporting information such as the term counts for the document, and the global weights used to compute the weights (if any). However, it does not have to use this information if it does not need to.

Parameters:

weights - The weights to normalize. Contains an input and should be modified to create an output. Cannot be null.

counts - The term counts for the document to normalize. Cannot be null.

globalWeights - The global weights to use in to create the weights, if any. Can be null.