info.ephyra.questionanalysis
Class Term

java.lang.Object
  info.ephyra.questionanalysis.Term

All Implemented Interfaces:

java.io.Serializable

public class Term
extends java.lang.Object
implements java.io.Serializable

A Term comprises one or more tokens of text that form a unit of meaning. It can be an individual word, a compound noun or a named entity.

This class implements the interface Serializable.

Version:

2008-01-23

Author:

Nico Schlaefer

See Also:

Serialized Form

Field Summary

static java.lang.String	COMPOUND Part of speech tag for terms that comprise multiple tokens
private java.util.Map<java.lang.String,java.lang.Double>	expansionLemmas Maps lemmas of the expansions to their weights.
private java.util.Map<java.lang.String,java.lang.Double>	expansions Maps expansions of the term to their weights.
private java.lang.String	lemma The lemma of the term.
private java.lang.String[]	neTypes The named entity types of the term (optional).
private java.lang.String	pos The part of speech of the term or COMPOUND to indicate that it comprises multiple tokens.
private double	relFrequency Relative frequency of the term.
private static long	serialVersionUID Version number used during deserialization.
private java.lang.String	text The textual representation of the term.

Constructor Summary

Term(java.lang.String text, java.lang.String pos)
Constructs a term from the provided information.

Term(java.lang.String text, java.lang.String pos, java.lang.String[] neTypes)
Constructs a term from the provided information.

Method Summary

private void	generateLemma() Generates the lemma of the term.
java.util.Map<java.lang.String,java.lang.Double>	getExpansions()
java.lang.String	getLemma()
java.lang.String[]	getNeTypes()
java.lang.String	getPos()
double	getRelFrequency()
java.lang.String	getText()
double	getWeight(java.lang.String lemma) Gets the weight of the term or expansion with the given lemma.
void	setExpansionLemmas(java.util.Map<java.lang.String,java.lang.Double> expansionLemmas) Normalizes and sets the lemmas of the expansions.
void	setExpansions(java.util.Map<java.lang.String,java.lang.Double> expansions)
void	setLemma(java.lang.String lemma) Normalizes and sets the lemma of the term.
void	setNeTypes(java.lang.String[] neTypes)
void	setRelFrequency(double relFrequency)
double	simScore(java.lang.String lemma) Calculates similarity scores for the given lemma and the lemmas of the term and its expansions based on their weights and the number of common tokens.
java.lang.String	toString() Creates a string representation of the term.

Methods inherited from class java.lang.Object

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

Field Detail

serialVersionUID

private static final long serialVersionUID

Version number used during deserialization.

See Also:

Constant Field Values

COMPOUND

public static final java.lang.String COMPOUND

Part of speech tag for terms that comprise multiple tokens

See Also:

Constant Field Values

text

private java.lang.String text

The textual representation of the term.

lemma

private java.lang.String lemma

The lemma of the term.

pos

private java.lang.String pos

The part of speech of the term or COMPOUND to indicate that it comprises multiple tokens.

neTypes

private java.lang.String[] neTypes

The named entity types of the term (optional).

relFrequency

private double relFrequency

Relative frequency of the term.

expansions

private java.util.Map<java.lang.String,java.lang.Double> expansions

Maps expansions of the term to their weights.

expansionLemmas

private java.util.Map<java.lang.String,java.lang.Double> expansionLemmas

Maps lemmas of the expansions to their weights.

Constructor Detail

Term

public Term(java.lang.String text,
            java.lang.String pos)

Constructs a term from the provided information.

Parameters:

text - textual representation

pos - part of speech

Term

public Term(java.lang.String text,
            java.lang.String pos,
            java.lang.String[] neTypes)

Constructs a term from the provided information.

Parameters:

text - textual representation

pos - part of speech

neTypes - named entity types

Method Detail

getText

public java.lang.String getText()

getLemma

public java.lang.String getLemma()

getPos

public java.lang.String getPos()

getNeTypes

public java.lang.String[] getNeTypes()

setNeTypes

public void setNeTypes(java.lang.String[] neTypes)

getRelFrequency

public double getRelFrequency()

setRelFrequency

public void setRelFrequency(double relFrequency)

getExpansions

public java.util.Map<java.lang.String,java.lang.Double> getExpansions()

setExpansions

public void setExpansions(java.util.Map<java.lang.String,java.lang.Double> expansions)

generateLemma

private void generateLemma()

Generates the lemma of the term.

setLemma

public void setLemma(java.lang.String lemma)

Normalizes and sets the lemma of the term.

Parameters:

lemma - the lemma of the term

setExpansionLemmas

public void setExpansionLemmas(java.util.Map<java.lang.String,java.lang.Double> expansionLemmas)

Normalizes and sets the lemmas of the expansions.

Parameters:

expansionLemmas - the lemmas of the expansions

getWeight

public double getWeight(java.lang.String lemma)

Gets the weight of the term or expansion with the given lemma.

Parameters:

lemma - the lemma

Returns:

the weight or 0 if there is no match

simScore

public double simScore(java.lang.String lemma)

Calculates similarity scores for the given lemma and the lemmas of the term and its expansions based on their weights and the number of common tokens. Gets the maximum of all these scores.

Parameters:

lemma - lemma to compare with

Returns:

similarity score

toString

public java.lang.String toString()

Creates a string representation of the term.

Overrides:

toString in class java.lang.Object

Returns:

string representation