Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

TEXT-126: Adding Sorensen-Dice similarity algorithm #621

Draft
wants to merge 4 commits into
base: master
Choose a base branch
from
Draft
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Original file line number Diff line number Diff line change
@@ -0,0 +1,139 @@
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package org.apache.commons.text.similarity;

import java.util.ArrayList;
import java.util.Collection;
import java.util.List;
import java.util.function.Function;

import org.apache.commons.lang3.StringUtils;

/**
* A similarity algorithm indicating the percentage of matched characters
* between two character sequences.
*
* <p>The S�rensen-Dice coefficient is a statistic used for comparing the
* similarity of two samples. It was independently developed by the botanists
* Thorvald S�rensen and Lee Raymond Dice, who published in 1948 and 1945
* respectively. The index is known by several other names, especially
* S�rensen-Dice index, S�rensen index and Dice's coefficient. Other
* variations include the "similarity coefficient" or "index", such as Dice
* similarity coefficient (DSC).</p>
*
* <p>This implementation is based on the S�rensen-Dice similarity algorithm
* from <a href=
* "https://en.wikipedia.org/wiki/Dice-S%C3%B8rensen_coefficient">
* https://en.wikipedia.org/wiki/Dice-S%C3%B8rensen_coefficient</a>.</p>
*
* @since 1.13
*/
public class SorensenDiceSimilarity implements SimilarityScore<Double> {

/**
* For shifting bigrams to fit in single integer.
*/
private static final int SHIFT_NUMBER = 16;

/**
* Converter function for conversion of string to bigrams.
*/
private final Function<CharSequence, Collection<Integer>> converter = new SorensenDiceConverter();

/**
* Measures the overlap of two sets created from a pair of character sequences.
* {@link IntersectionSimilarity}}
*/
private final IntersectionSimilarity<Integer> similarity = new IntersectionSimilarity<>(this.converter);

/**
* Calculates Sorensen-Dice Similarity of two character sequences passed as
* input.
*
* <pre>
* similarity.apply(null, null) = IllegalArgumentException
* similarity.apply("foo", null) = IllegalArgumentException
* similarity.apply(null, "foo") = IllegalArgumentException
* similarity.apply("night", "nacht") = 0.25
* similarity.apply("", "") = 1.0
* similarity.apply("foo", "foo") = 1.0
* similarity.apply("foo", "foo ") = 0.8
* similarity.apply("foo", " foo") = 0.8
* similarity.apply("foo", " foo ") &asymp; 0.66&hellip;
* similarity.apply("", "a") = 0.0
* similarity.apply("aaapppp", "") = 0.0
* similarity.apply("frog", "fog") = 0.4
* similarity.apply("fly", "ant") = 0.0
* similarity.apply("elephant", "hippo") = 0.0
* similarity.apply("hippo", "elephant") = 0.0
* similarity.apply("hippo", "zzzzzzzz") = 0.0
* similarity.apply("hello", "hallo") = 0.5
* similarity.apply("ABC Corporation", "ABC Corp") &asymp; 0.66&hellip;
* similarity.apply("D N H Enterprises Inc", "D &amp; H Enterprises, Inc.") = 0.74
* similarity.apply("My Gym Children's Fitness Center", "My Gym. Childrens Fitness") = 0.76
* similarity.apply("PENNSYLVANIA", "PENNCISYLVNIA") = 0.69
* </pre>
*
* @param left the first CharSequence, must not be null
* @param right the second CharSequence, must not be null
* @return result similarity
* @throws IllegalArgumentException if either CharSequence input is {@code null}
*/
@Override
public Double apply(final CharSequence left, final CharSequence right) {

if (left == null || right == null) {
throw new IllegalArgumentException("CharSequences must not be null");
}

if (StringUtils.equals(left, right)) {
return 1d;
}

// if bigram is not formed out of any given string, clearly both are not similar.
if (left.length() < 2 || right.length() < 2) {
return 0d;
}

IntersectionResult overlap = similarity.apply(left, right);

final int total = overlap.getSizeA() + overlap.getSizeB();
final long intersection = overlap.getIntersection();

return (2.0d * intersection) / total;
}

/**
* Converter class for creating Bigrams for SorensenDice similarity.
*/
private static final class SorensenDiceConverter implements Function<CharSequence, Collection<Integer>> {
@Override
public Collection<Integer> apply(CharSequence cs) {
final int length = cs.length();
final List<Integer> list = new ArrayList<>(length);
if (length > 1) {
char ch2 = cs.charAt(0);
for (int i = 1; i < length; i++) {
final char ch1 = ch2;
ch2 = cs.charAt(i);
list.add((ch1 << SHIFT_NUMBER) | ch2);
}
}
return list;
}
}
}
Original file line number Diff line number Diff line change
@@ -0,0 +1,99 @@
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package org.apache.commons.text.similarity;

import static org.junit.jupiter.api.Assertions.assertEquals;
import static org.junit.jupiter.api.Assertions.assertThrows;

import java.math.RoundingMode;
import java.text.DecimalFormat;

import org.junit.jupiter.api.BeforeAll;
import org.junit.jupiter.api.Test;

/**
* Unit tests for {@link SorensenDiceSimilarity}.
*/
public class SorensenDiceSimilarityTest {

private static SorensenDiceSimilarity similarity;

@BeforeAll
public static void setUp() {
similarity = new SorensenDiceSimilarity();
}

@Test
public void test() {
assertEquals(0.25d, similarity.apply("night", "nacht"));
}

@Test
public void testGetSorensenDiceSimilarity_StringString() {
assertEquals(1d, similarity.apply("", ""));
assertEquals(0d, similarity.apply("", "a"));
assertEquals(0d, similarity.apply("a", ""));
assertEquals(1d, similarity.apply("a", "a"));
assertEquals(0d, similarity.apply("a", "b"));
assertEquals(1.0d, similarity.apply("foo", "foo"));
assertEquals(0.8d, similarity.apply("foo", "foo "));
assertEquals(0.8d, similarity.apply("foo", " foo"));
assertEquals(0.66d, printTwoDecimals(similarity.apply("foo", " foo ")));
assertEquals(0.4d, similarity.apply("frog", "fog"));
assertEquals(0.0d, similarity.apply("fly", "ant"));
assertEquals(0.0d, similarity.apply("elephant", "hippo"));
assertEquals(0.0d, similarity.apply("hippo", "elephant"));
assertEquals(0.0d, similarity.apply("hippo", "zzzzzzzz"));
assertEquals(0.25d, similarity.apply("night", "nacht"));
assertEquals(0.5d, similarity.apply("hello", "hallo"));
assertEquals(0.0d, similarity.apply("aaapppp", ""));
assertEquals(0.66d, printTwoDecimals(similarity.apply("ABC Corporation", "ABC Corp")));
assertEquals(0.73d, printTwoDecimals(similarity.apply("D N H Enterprises Inc", "D &amp; H Enterprises, Inc.")));
assertEquals(0.76d, printTwoDecimals(similarity.apply("My Gym Children's Fitness Center", "My Gym. Childrens Fitness")));
assertEquals(0.69d, printTwoDecimals(similarity.apply("PENNSYLVANIA", "PENNCISYLVNIA")));
assertEquals(0.92d, printTwoDecimals(similarity.apply("/opt/software1", "/opt/software2")));
assertEquals(0.60d, printTwoDecimals(similarity.apply("aaabcd", "aaacdb")));
assertEquals(0.63d, printTwoDecimals(similarity.apply("John Horn", "John Hopkins")));
}

@Test
public void testGetSorensenDicesSimilarity_NullNull() {
assertThrows(IllegalArgumentException.class, () -> similarity.apply(null, null));
}

@Test
public void testGetSorensenDicesSimilarity_StringNull() {
assertThrows(IllegalArgumentException.class, () -> similarity.apply(" ", null));
}

@Test
public void testGetSorensenDicesSimilarity_NullString() {
assertThrows(IllegalArgumentException.class, () -> similarity.apply(null, "clear"));
}

/**
* Format the double to two decimal places rounding it down.
*
* @param value the double value
* @return double formatted to two places rounding it down
*/
public static double printTwoDecimals(double value) {
DecimalFormat df = new DecimalFormat("#.##");
df.setRoundingMode(RoundingMode.DOWN);
return Double.parseDouble(df.format(value));
}
}