edu.ucla.sspace.common

Class DocumentVectorBuilder

java.lang.Object

edu.ucla.sspace.common.DocumentVectorBuilder

public class DocumentVectorBuilder
extends Object

DocumentVectorBuilder generates Vector representations of a document, based on semantic Vectors provided for a SemanticSpace. This can be consider as a projecting the document into the semantic space.

Documents will be tokenized using the current tokenizing method, and the vector in the SemanticSpace corresponding to each word found in the document will be combined together.

Options for combining term Vectors include summation, average, and term frequency weighting.

Author:

Keith Stevens

Field Summary

Fields

Modifier and Type	Field and Description
static String	USE_TERM_FREQUENCIES_PROPERTY The property to specify if term frequencies should be used when combining term vectors.

Constructor Summary

Constructors

Constructor and Description
DocumentVectorBuilder(SemanticSpace baseSpace) Creates a DocumentVectorBuilder from a SemanticSpace and extracts options from the system wide Properties.
DocumentVectorBuilder(SemanticSpace baseSpace, Properties props) Creates a DocumentVectorBuilder from a SemanticSpace and extracts options from the given Properties.

Method Summary

Methods

Modifier and Type	Method and Description
void	add(DoubleVector dest, Vector src, int factor)
DoubleVector	buildVector(BufferedReader document, DoubleVector documentVector) Represent a document as the summation of term Vectors.

Methods inherited from class java.lang.Object

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

Field Detail

USE_TERM_FREQUENCIES_PROPERTY

public static final String USE_TERM_FREQUENCIES_PROPERTY

The property to specify if term frequencies should be used when combining term vectors.

See Also:

Constant Field Values

Constructor Detail

DocumentVectorBuilder

public DocumentVectorBuilder(SemanticSpace baseSpace)

Creates a DocumentVectorBuilder from a SemanticSpace and extracts options from the system wide Properties.

DocumentVectorBuilder

public DocumentVectorBuilder(SemanticSpace baseSpace,
                     Properties props)

Creates a DocumentVectorBuilder from a SemanticSpace and extracts options from the given Properties.

Method Detail

buildVector

public DoubleVector buildVector(BufferedReader document,
                       DoubleVector documentVector)

Represent a document as the summation of term Vectors.

Parameters:

document - A BufferedReader for a document to project into a SemanticSpace.

documentVector - A Vector which has been pre-allocated to store the document's representation. This is pre-allocated so that users of DocumentVectorBuilder can decide what type of Vector should be used to represent a document.

Returns:

documentVector after it has been modified to represent the terms in document.

add

public void add(DoubleVector dest,
       Vector src,
       int factor)