org.carrot2.text.vsm
Class VectorSpaceModelContext

java.lang.Object
  org.carrot2.text.vsm.VectorSpaceModelContext

public class VectorSpaceModelContext
extends Object

Stores data related to the Vector Space Model of the processed documents.

Field Summary

PreprocessingContext	preprocessingContext Preprocessing context for the underlying documents.
com.carrotsearch.hppc.IntIntOpenHashMap	stemToRowIndex Stem index to row index mapping for the tdMatrix.
org.apache.mahout.math.matrix.DoubleMatrix2D	termDocumentMatrix Term-document matrix.
org.apache.mahout.math.matrix.DoubleMatrix2D	termPhraseMatrix Term-document-like matrix for phrases from PreprocessingContext.AllLabels.

Constructor Summary

VectorSpaceModelContext(PreprocessingContext preprocessingContext)
Creates a vector space model context with the provided preprocessing context.

Method Summary

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

preprocessingContext

public final PreprocessingContext preprocessingContext

Preprocessing context for the underlying documents.

termDocumentMatrix

public org.apache.mahout.math.matrix.DoubleMatrix2D termDocumentMatrix

Term-document matrix. Rows of the matrix correspond to word stems, columns correspond to the processed documents. For mapping between rows of this matrix and PreprocessingContext.AllStems, see stemToRowIndex.

This matrix is produced by TermDocumentMatrixBuilder.buildTermDocumentMatrix(VectorSpaceModelContext).

termPhraseMatrix

public org.apache.mahout.math.matrix.DoubleMatrix2D termPhraseMatrix

Term-document-like matrix for phrases from PreprocessingContext.AllLabels. If there are no phrases in PreprocessingContext.AllLabels, phrase matrix is null. For mapping between rows of this matrix and PreprocessingContext.AllStems, see stemToRowIndex.

This matrix is produced by TermDocumentMatrixBuilder.buildTermPhraseMatrix(VectorSpaceModelContext).

stemToRowIndex

public com.carrotsearch.hppc.IntIntOpenHashMap stemToRowIndex

Stem index to row index mapping for the tdMatrix. Keys in this map are indices of entries in PreprocessingContext.AllStems arrays, values are the indices of tdMatrix rows corresponding to the stems. Please note that depending on the limit on the size of the matrix, some stems may not have their corresponding matrix rows.

This object is produced by TermDocumentMatrixBuilder.buildTermDocumentMatrix(VectorSpaceModelContext).

Constructor Detail

VectorSpaceModelContext

public VectorSpaceModelContext(PreprocessingContext preprocessingContext)

Creates a vector space model context with the provided preprocessing context.