de.hpi.fgis.dude.similarityfunction.contentbased.util
Class MongeElkanSimilarity

java.lang.Object
  uk.ac.shef.wit.simmetrics.similaritymetrics.AbstractStringMetric
      de.hpi.fgis.dude.similarityfunction.contentbased.util.MongeElkanSimilarity

All Implemented Interfaces:

Serializable, uk.ac.shef.wit.simmetrics.similaritymetrics.InterfaceStringMetric

public class MongeElkanSimilarity
extends uk.ac.shef.wit.simmetrics.similaritymetrics.AbstractStringMetric
implements Serializable

This class is a derived implementation of the Monge Elkan metric from the simmetric library. Internal String Comparison was changed to SmithWaterman. Package: uk.ac.shef.wit.simmetrics.similaritymetrics.mongeelkan Description: uk.ac.shef.wit.simmetrics.similaritymetrics.mongeelkan implements a

See Also:

Serialized Form

Field Summary

(package private) uk.ac.shef.wit.simmetrics.tokenisers.InterfaceTokeniser

tokeniser private tokeniser for tokenisation of the query strings.

Constructor Summary

MongeElkanSimilarity()
constructor - default (empty).

MongeElkanSimilarity(uk.ac.shef.wit.simmetrics.similaritymetrics.AbstractStringMetric metricToUse)
constructor.

MongeElkanSimilarity(uk.ac.shef.wit.simmetrics.tokenisers.InterfaceTokeniser tokeniserToUse)
constructor.

MongeElkanSimilarity(uk.ac.shef.wit.simmetrics.tokenisers.InterfaceTokeniser tokeniserToUse, uk.ac.shef.wit.simmetrics.similaritymetrics.AbstractStringMetric metricToUse)
constructor.

Method Summary

String	getLongDescriptionString() returns the long string identifier for the metric.
String	getShortDescriptionString() returns the string identifier for the metric.
float	getSimilarity(String string1, String string2) gets the similarity of the two strings using Monge Elkan.
String	getSimilarityExplained(String string1, String string2) gets a div class xhtml similarity explaining the operation of the metric.
float	getSimilarityTimingEstimated(String string1, String string2) gets the estimated time in milliseconds it takes to perform a similarity timing.
float	getUnNormalisedSimilarity(String string1, String string2) gets the un-normalised similarity measure of the metric for the given strings.

Methods inherited from class uk.ac.shef.wit.simmetrics.similaritymetrics.AbstractStringMetric

batchCompareSet, batchCompareSets, getSimilarityTimingActual

Methods inherited from class java.lang.Object

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

Field Detail

tokeniser

final uk.ac.shef.wit.simmetrics.tokenisers.InterfaceTokeniser tokeniser

private tokeniser for tokenisation of the query strings.

Constructor Detail

MongeElkanSimilarity

public MongeElkanSimilarity()

constructor - default (empty).

MongeElkanSimilarity

public MongeElkanSimilarity(uk.ac.shef.wit.simmetrics.tokenisers.InterfaceTokeniser tokeniserToUse)

constructor.

Parameters:

tokeniserToUse - - the tokeniser to use should a different tokeniser be required

MongeElkanSimilarity

public MongeElkanSimilarity(uk.ac.shef.wit.simmetrics.tokenisers.InterfaceTokeniser tokeniserToUse,
                            uk.ac.shef.wit.simmetrics.similaritymetrics.AbstractStringMetric metricToUse)

constructor.

Parameters:

tokeniserToUse - - the tokeniser to use should a different tokeniser be required

metricToUse - - the string metric to use

MongeElkanSimilarity

public MongeElkanSimilarity(uk.ac.shef.wit.simmetrics.similaritymetrics.AbstractStringMetric metricToUse)

constructor.

Parameters:

metricToUse - - the string metric to use

Method Detail

getSimilarityExplained

public String getSimilarityExplained(String string1,
                                     String string2)

gets a div class xhtml similarity explaining the operation of the metric.

Specified by:

getSimilarityExplained in interface uk.ac.shef.wit.simmetrics.similaritymetrics.InterfaceStringMetric

Specified by:

getSimilarityExplained in class uk.ac.shef.wit.simmetrics.similaritymetrics.AbstractStringMetric

Parameters:

string1 - string 1

string2 - string 2

Returns:

a div class html section detailing the metric operation.

getShortDescriptionString

public String getShortDescriptionString()

returns the string identifier for the metric.

Specified by:

getShortDescriptionString in interface uk.ac.shef.wit.simmetrics.similaritymetrics.InterfaceStringMetric

Specified by:

getShortDescriptionString in class uk.ac.shef.wit.simmetrics.similaritymetrics.AbstractStringMetric

Returns:

the string identifier for the metric

getLongDescriptionString

public String getLongDescriptionString()

returns the long string identifier for the metric.

Specified by:

getLongDescriptionString in interface uk.ac.shef.wit.simmetrics.similaritymetrics.InterfaceStringMetric

Specified by:

getLongDescriptionString in class uk.ac.shef.wit.simmetrics.similaritymetrics.AbstractStringMetric

Returns:

the long string identifier for the metric

getSimilarityTimingEstimated

public float getSimilarityTimingEstimated(String string1,
                                          String string2)

gets the estimated time in milliseconds it takes to perform a similarity timing.

Specified by:

getSimilarityTimingEstimated in interface uk.ac.shef.wit.simmetrics.similaritymetrics.InterfaceStringMetric

Specified by:

getSimilarityTimingEstimated in class uk.ac.shef.wit.simmetrics.similaritymetrics.AbstractStringMetric

Parameters:

string1 - string 1

string2 - string 2

Returns:

the estimated time in milliseconds taken to perform the similarity measure

getSimilarity

public final float getSimilarity(String string1,
                                 String string2)

gets the similarity of the two strings using Monge Elkan.

Specified by:

getSimilarity in interface uk.ac.shef.wit.simmetrics.similaritymetrics.InterfaceStringMetric

Specified by:

getSimilarity in class uk.ac.shef.wit.simmetrics.similaritymetrics.AbstractStringMetric

Parameters:

string1 -

string2 -

Returns:

a value between 0-1 of the similarity

getUnNormalisedSimilarity

public float getUnNormalisedSimilarity(String string1,
                                       String string2)

gets the un-normalised similarity measure of the metric for the given strings.

Specified by:

getUnNormalisedSimilarity in class uk.ac.shef.wit.simmetrics.similaritymetrics.AbstractStringMetric

Parameters:

string1 -

string2 -

Returns:

returns the score of the similarity measure (un-normalised)