com.wcohen.secondstring
Class SLIM

java.lang.Object
  |
  +--com.wcohen.secondstring.AbstractStringDistance
        |
        +--com.wcohen.secondstring.SLIM

All Implemented Interfaces:

StringDistance

public class SLIM

extends AbstractStringDistance

The same-letter index mixture distance.

Experimental, invented by William Cohen. Ahem. Ahem. Which is mine.

Represents a pair S=a1...aK, T=b1...bK as a set of index distances D = { i-j : ai=bj }, and models this set D as a mixture of two Gaussians, one with small variance (so i~=j) and one with large variance (so i!=j). The degree of overlap for two chars ai and bj such that ai=bj is the posterior probability of the small-variance Gaussian, given i-j. Distance is average overlap between the two strings.

Constructor Summary

SLIM()

Method Summary

java.lang.String	explainScore(StringWrapper s, StringWrapper t) This method needs to be implemented by subclasses.
static void	main(java.lang.String[] argv)
StringWrapper	prepare(java.lang.String s) Default way to preprocess a string for distance computation.
double	score(StringWrapper s, StringWrapper t) This method needs to be implemented by subclasses.
java.lang.String	toString()

Methods inherited from class com.wcohen.secondstring.AbstractStringDistance

accumulateStatistics, doMain, explainScore, score

Methods inherited from class java.lang.Object

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

Constructor Detail

SLIM

public SLIM()

Method Detail

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object

score

public double score(StringWrapper s,
                    StringWrapper t)

Description copied from class: AbstractStringDistance

This method needs to be implemented by subclasses.

Specified by:

score in interface StringDistance

Specified by:

score in class AbstractStringDistance

explainScore

public java.lang.String explainScore(StringWrapper s,
                                     StringWrapper t)

Description copied from class: AbstractStringDistance

This method needs to be implemented by subclasses.

Specified by:

explainScore in interface StringDistance

Specified by:

explainScore in class AbstractStringDistance

prepare

public StringWrapper prepare(java.lang.String s)

Description copied from class: AbstractStringDistance

Default way to preprocess a string for distance computation. If this is an expensive operations, then override this method to return a StringWrapper implementation that caches appropriate information about s.

Specified by:

prepare in interface StringDistance

Overrides:

prepare in class AbstractStringDistance

main

public static void main(java.lang.String[] argv)