edu.umass.cs.mallet.base.types

Class StringKernel

java.lang.Object

java.util.AbstractMap<K,V>

java.util.HashMap<K,V>

java.util.LinkedHashMap

edu.umass.cs.mallet.base.types.StringKernel

All Implemented Interfaces:

Serializable, Cloneable, Map

public class StringKernel
extends LinkedHashMap

Computes a similarity metric between two strings, based on counts of common subsequences of characters. See Lodhi et al "String kernels for text classification." Optionally caches previous kernel computations.

See Also:

Serialized Form

Nested Class Summary

Nested classes/interfaces inherited from class java.util.AbstractMap

AbstractMap.SimpleEntry<K,V>, AbstractMap.SimpleImmutableEntry<K,V>

Constructor Summary

Constructors

Constructor and Description
StringKernel()
StringKernel(boolean norm, double lam, int length)
StringKernel(boolean norm, double lam, int length, boolean cache)

Method Summary

Modifier and Type	Method and Description
double	K(String s, String t) Computes the normalized string kernel between two strings.
static void	main(String[] args) Return string kernel between two strings

Methods inherited from class java.util.LinkedHashMap

clear, containsValue, entrySet, forEach, get, getOrDefault, keySet, removeEldestEntry, replaceAll, values

Methods inherited from class java.util.HashMap

clone, compute, computeIfAbsent, computeIfPresent, containsKey, isEmpty, merge, put, putAll, putIfAbsent, remove, remove, replace, replace, size

Methods inherited from class java.util.AbstractMap

equals, hashCode, toString

Methods inherited from class java.lang.Object

finalize, getClass, notify, notifyAll, wait, wait, wait

Methods inherited from interface java.util.Map

compute, computeIfAbsent, computeIfPresent, containsKey, equals, hashCode, isEmpty, merge, put, putAll, putIfAbsent, remove, remove, replace, replace, size

Constructor Detail

StringKernel

public StringKernel(boolean norm,
                    double lam,
                    int length,
                    boolean cache)

Parameters:

norm - true if we lowercase all strings

lam - 0-1 penalty for gaps between matches.

length - max length of subsequences to compare

cache - true if we should cache previous kernel computations. recommended!

StringKernel

public StringKernel()

StringKernel

public StringKernel(boolean norm,
                    double lam,
                    int length)

Method Detail

K

public double K(String s,
                String t)

Computes the normalized string kernel between two strings.

Parameters:

s - string 1

t - string 2

Returns:

0-1 value, where 1 is exact match.

main

public static void main(String[] args)
                 throws Exception

Return string kernel between two strings

Throws:

Exception