cat.lump.ir.weighting

Class TermFrequency

java.lang.Object

cat.lump.ir.weighting.TermFrequency

public class TermFrequency
extends java.lang.Object

A class to compute and store a simple term frequency. It includes the methods to add and remove terms, get the sorted list of TermFrequencyTuples, and just a subset of them

Author:

albarron

Constructor Summary

Constructors

Constructor and Description
TermFrequency() Invokes the class with an empty list of term tuples
TermFrequency(java.util.List<TermFrequencyTuple> terms) Invokes the class with an existing empty list of term tuples

Method Summary

Methods

Modifier and Type	Method and Description
void	addTerm(java.lang.String term) Add a term into the collection.
void	addTerms(java.util.List<java.lang.String> terms) Add these terms into the collection.
boolean	existTerm(java.lang.String term)
java.util.List<TermFrequencyTuple>	getAll()
TermFrequencyTuple	getTerm(java.lang.String term)
java.util.List<TermFrequencyTuple>	getTop(int top, int max) Subset of terms with the highest tf up to top% or up to max Note that not the top% is returned sometimes but a little bit more.
java.util.List<TermFrequencyTuple>	getTopPlus(int top, int max, java.util.List<java.lang.String> catTerms) Note that not the top% is returned sometimes but a little bit more.
void	removeTerm(java.lang.String term) Remove the given term (warning if the term does not exist)
void	removeTerms(java.util.List<java.lang.String> terms) Remove these terms into the collection.
int	size()

Methods inherited from class java.lang.Object

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

Constructor Detail

TermFrequency

public TermFrequency()

Invokes the class with an empty list of term tuples

TermFrequency

public TermFrequency(java.util.List<TermFrequencyTuple> terms)

Invokes the class with an existing empty list of term tuples

Parameters:

terms -

Method Detail

existTerm

public boolean existTerm(java.lang.String term)

Parameters:

term -

Returns:

true if the term exists

addTerm

public void addTerm(java.lang.String term)

Add a term into the collection. The frequency of the term is increased if it already exists. Otherwise, it is set to 1.

Parameters:

term - a non-null, non-empty string

removeTerm

public void removeTerm(java.lang.String term)

Remove the given term (warning if the term does not exist)

Parameters:

term - non-null string

addTerms

public void addTerms(java.util.List<java.lang.String> terms)

Add these terms into the collection.

Parameters:

terms -

removeTerms

public void removeTerms(java.util.List<java.lang.String> terms)

Remove these terms into the collection.

Parameters:

terms -

getTerm

public TermFrequencyTuple getTerm(java.lang.String term)

Parameters:

term -

Returns:

the term and its weight

getAll

public java.util.List<TermFrequencyTuple> getAll()

Returns:

The list of term frequency tuples. This tuples are sorted by frequency in descending order.

getTop

public java.util.List<TermFrequencyTuple> getTop(int top,
                                        int max)

Subset of terms with the highest tf up to top% or up to max Note that not the top% is returned sometimes but a little bit more. It happens because the collection is increased until the last considered frequency drops.

Parameters:

top - percentage of the terms to get (0 < top <=100)

maximum - number of terms allowed in the vocabulary (max > 1) max=-1 indicates no maximum value

Returns:

subset of the terms with the highest tf up to top%

getTopPlus

public java.util.List<TermFrequencyTuple> getTopPlus(int top,
                                            int max,
                                            java.util.List<java.lang.String> catTerms)

Note that not the top% is returned sometimes but a little bit more. It happens because the collection is increased until the last considered frequency drops.

Parameters:

top - percentage of the terms to get (0 < top <=100)

Returns:

subset of the terms with the highest tf up to top% plus the terms that appear in the title of the category

size

public int size()