edu.ucla.sspace.coals

Class Coals

java.lang.Object

edu.ucla.sspace.coals.Coals

All Implemented Interfaces:

SemanticSpace

public class Coals
extends Object
implements SemanticSpace

An implementation of the COALS Semantic Space model. This implementation is based on:

Rohde, D. L. T., Gonnerman, L. M., Plaut, D. C. (2005). An Improved Model of Semantic Similarity Based on Lexical Co-Occurrence. Cognitive Science (submitted). Available here

COALS first computes a term by term co-occurrance using a ramped 4-word window. Once all documents have been processed, the co-occurrence matrix will be re ordered such that only the N most frequent terms have their semantic vectors retained and only the M most frequent terms are used as co-occurrence features. These values can be set by the "edu.ucla.sspace.coals.Coals.maxWords" and "edu.ucla.sspace.coals.Coals.maxDimensions" properties, resepctively. After re ordering the semantic vectors and features, CorrelationTransform is used to rerank all co-occurrence scores. As part of this transform, all negative correlations are dropped and replaced with a 0. Finally, and optionally, the SVD is used to reduce the semantic space. To set the number of retained dimensions via SVD, set the "edu.ucla.sspace.coals.Coals.dimension" property.

Author:

Keith Stevens

Field Summary

Fields

Modifier and Type	Field and Description
static String	COALS_SSPACE_NAME The name of this SemanticSpace
static String	DO_NOT_NORMALIZE_PROPERTY Specifies if Coals should not normalize the co-occurance matrix.
static String	MAX_DIMENSIONS_PROPERTY Specifies the number of dimensions in the raw co-occurrance matrix to maintain.
static String	MAX_WORDS_PROPERTY Specifies the number of words to build semantics for.
static String	PROPERTY_PREFIX The property prefix for other settings.
static String	REDUCE_DIMENSION_PROPERTY Specifies the number of dimensions the co-occurance matrix should be reduced to.
static String	REDUCE_MATRIX_PROPERTY Specifies whether or not the co-occurance matrix should be reduced.

Constructor Summary

Constructors

Constructor and Description
Coals(Transform transform, MatrixFactorization reducer)
Coals(Transform transform, MatrixFactorization reducer, int reducedDimensions, int maxWords, int maxDimensions) Creats a Coals instance.

Method Summary

Methods

Modifier and Type	Method and Description
String	getSpaceName() Returns a unique string describing the name and configuration of this algorithm.
Vector	getVector(String term) Returns the semantic vector for the provided word.
int	getVectorLength() Returns the length of vectors in this semantic space.
Set<String>	getWords() Returns the set of words that are represented in this semantic space.
void	processDocument(BufferedReader document) Processes the contents of the provided file as a document.
void	processSpace(Properties props) Once all the documents have been processed, performs any post-processing steps on the data.

Methods inherited from class java.lang.Object

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

Field Detail

PROPERTY_PREFIX

public static final String PROPERTY_PREFIX

The property prefix for other settings.

See Also:

Constant Field Values

REDUCE_MATRIX_PROPERTY

public static final String REDUCE_MATRIX_PROPERTY

Specifies whether or not the co-occurance matrix should be reduced.

See Also:

Constant Field Values

REDUCE_DIMENSION_PROPERTY

public static final String REDUCE_DIMENSION_PROPERTY

Specifies the number of dimensions the co-occurance matrix should be reduced to.

See Also:

Constant Field Values

MAX_DIMENSIONS_PROPERTY

public static final String MAX_DIMENSIONS_PROPERTY

Specifies the number of dimensions in the raw co-occurrance matrix to maintain.

See Also:

Constant Field Values

MAX_WORDS_PROPERTY

public static final String MAX_WORDS_PROPERTY

Specifies the number of words to build semantics for.

See Also:

Constant Field Values

DO_NOT_NORMALIZE_PROPERTY

public static final String DO_NOT_NORMALIZE_PROPERTY

Specifies if Coals should not normalize the co-occurance matrix.

See Also:

Constant Field Values

COALS_SSPACE_NAME

public static final String COALS_SSPACE_NAME

The name of this SemanticSpace

See Also:

Constant Field Values

Constructor Detail

Coals

public Coals(Transform transform,
     MatrixFactorization reducer)

Coals

public Coals(Transform transform,
     MatrixFactorization reducer,
     int reducedDimensions,
     int maxWords,
     int maxDimensions)

Creats a Coals instance.

Method Detail

getWords

public Set<String> getWords()

Returns the set of words that are represented in this semantic space.

Specified by:

getWords in interface SemanticSpace

Returns:

the set of words that are represented in this semantic space.

getVector

public Vector getVector(String term)

Returns the semantic vector for the provided word.

Specified by:

getVector in interface SemanticSpace

Parameters:

term - a word that may be in the semantic space

Returns:

The Vector for the provided word or null if the word was not in the space.

getSpaceName

public String getSpaceName()

Description copied from interface: SemanticSpace

Returns a unique string describing the name and configuration of this algorithm. Any configurable parameters that would affect the resulting semantic space should be expressed as a part of this name.

Specified by:

getSpaceName in interface SemanticSpace

getVectorLength

public int getVectorLength()

Description copied from interface: SemanticSpace

Returns the length of vectors in this semantic space. Implementations are left free to define whether the returned value is valid before processSpace is called.

Specified by:

getVectorLength in interface SemanticSpace

processDocument

public void processDocument(BufferedReader document)
                     throws IOException

Processes the contents of the provided file as a document.

Specified by:

processDocument in interface SemanticSpace

Parameters:

document - a reader that allows access to the text of the document

Throws:

IOException - if any error occurs while reading the document

processSpace

public void processSpace(Properties props)

Once all the documents have been processed, performs any post-processing steps on the data. An algorithm should treat this as a no-op if no post-processing is required. Callers may specify the values for any exposed parameters using the properties argument.

By general contract, once this method has been called, processDocument will not be called again.

Specified by:

processSpace in interface SemanticSpace

Parameters:

props - a set of properties and values that may be used to configure any exposed parameters of the algorithm.