001/*
002 * Licensed to the Apache Software Foundation (ASF) under one or more
003 * contributor license agreements.  See the NOTICE file distributed with
004 * this work for additional information regarding copyright ownership.
005 * The ASF licenses this file to You under the Apache License, Version 2.0
006 * (the "License"); you may not use this file except in compliance with
007 * the License.  You may obtain a copy of the License at
008 *
009 *      http://www.apache.org/licenses/LICENSE-2.0
010 *
011 * Unless required by applicable law or agreed to in writing, software
012 * distributed under the License is distributed on an "AS IS" BASIS,
013 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
014 * See the License for the specific language governing permissions and
015 * limitations under the License.
016 */
017package org.apache.commons.text.similarity;
018
019import java.util.Arrays;
020
021/**
022 * A similarity algorithm indicating the percentage of matched characters between two character sequences.
023 *
024 * <p>
025 * The Jaro measure is the weighted sum of percentage of matched characters
026 * from each file and transposed characters. Winkler increased this measure
027 * for matching initial characters.
028 * </p>
029 *
030 * <p>
031 * This implementation is based on the Jaro Winkler similarity algorithm
032 * from <a href="http://en.wikipedia.org/wiki/Jaro%E2%80%93Winkler_distance">
033 * http://en.wikipedia.org/wiki/Jaro%E2%80%93Winkler_distance</a>.
034 * </p>
035 *
036 * <p>
037 * This code has been adapted from Apache Commons Lang 3.3.
038 * </p>
039 *
040 * @since 1.0
041 */
042public class JaroWinklerDistance implements SimilarityScore<Double> {
043
044    /**
045     * The default prefix length limit set to four.
046     */
047    private static final int PREFIX_LENGTH_LIMIT = 4;
048    /**
049     * Represents a failed index search.
050     */
051    public static final int INDEX_NOT_FOUND = -1;
052
053    /**
054     * Find the Jaro Winkler Distance which indicates the similarity score
055     * between two CharSequences.
056     *
057     * <pre>
058     * distance.apply(null, null)          = IllegalArgumentException
059     * distance.apply("","")               = 0.0
060     * distance.apply("","a")              = 0.0
061     * distance.apply("aaapppp", "")       = 0.0
062     * distance.apply("frog", "fog")       = 0.93
063     * distance.apply("fly", "ant")        = 0.0
064     * distance.apply("elephant", "hippo") = 0.44
065     * distance.apply("hippo", "elephant") = 0.44
066     * distance.apply("hippo", "zzzzzzzz") = 0.0
067     * distance.apply("hello", "hallo")    = 0.88
068     * distance.apply("ABC Corporation", "ABC Corp") = 0.93
069     * distance.apply("D N H Enterprises Inc", "D &amp; H Enterprises, Inc.") = 0.95
070     * distance.apply("My Gym Children's Fitness Center", "My Gym. Childrens Fitness") = 0.92
071     * distance.apply("PENNSYLVANIA", "PENNCISYLVNIA")    = 0.88
072     * </pre>
073     *
074     * @param left the first String, must not be null
075     * @param right the second String, must not be null
076     * @return result distance
077     * @throws IllegalArgumentException if either String input {@code null}
078     */
079    @Override
080    public Double apply(final CharSequence left, final CharSequence right) {
081        final double defaultScalingFactor = 0.1;
082        final double percentageRoundValue = 100.0;
083
084        if (left == null || right == null) {
085            throw new IllegalArgumentException("Strings must not be null");
086        }
087
088        int[] mtp = matches(left, right);
089        double m = mtp[0];
090        if (m == 0) {
091            return 0D;
092        }
093        double j = ((m / left.length() + m / right.length() + (m - mtp[1]) / m)) / 3;
094        double jw = j < 0.7D ? j : j + Math.min(defaultScalingFactor, 1D / mtp[3]) * mtp[2] * (1D - j);
095        return Math.round(jw * percentageRoundValue) / percentageRoundValue;
096    }
097
098    /**
099     * This method returns the Jaro-Winkler string matches, transpositions, prefix, max array.
100     *
101     * @param first the first string to be matched
102     * @param second the second string to be machted
103     * @return mtp array containing: matches, transpositions, prefix, and max length
104     */
105    protected static int[] matches(final CharSequence first, final CharSequence second) {
106        CharSequence max, min;
107        if (first.length() > second.length()) {
108            max = first;
109            min = second;
110        } else {
111            max = second;
112            min = first;
113        }
114        int range = Math.max(max.length() / 2 - 1, 0);
115        int[] matchIndexes = new int[min.length()];
116        Arrays.fill(matchIndexes, -1);
117        boolean[] matchFlags = new boolean[max.length()];
118        int matches = 0;
119        for (int mi = 0; mi < min.length(); mi++) {
120            char c1 = min.charAt(mi);
121            for (int xi = Math.max(mi - range, 0), xn = Math.min(mi + range + 1, max.length()); xi < xn; xi++) {
122                if (!matchFlags[xi] && c1 == max.charAt(xi)) {
123                    matchIndexes[mi] = xi;
124                    matchFlags[xi] = true;
125                    matches++;
126                    break;
127                }
128            }
129        }
130        char[] ms1 = new char[matches];
131        char[] ms2 = new char[matches];
132        for (int i = 0, si = 0; i < min.length(); i++) {
133            if (matchIndexes[i] != -1) {
134                ms1[si] = min.charAt(i);
135                si++;
136            }
137        }
138        for (int i = 0, si = 0; i < max.length(); i++) {
139            if (matchFlags[i]) {
140                ms2[si] = max.charAt(i);
141                si++;
142            }
143        }
144        int transpositions = 0;
145        for (int mi = 0; mi < ms1.length; mi++) {
146            if (ms1[mi] != ms2[mi]) {
147                transpositions++;
148            }
149        }
150        int prefix = 0;
151        for (int mi = 0; mi < min.length(); mi++) {
152            if (first.charAt(mi) == second.charAt(mi)) {
153                prefix++;
154            } else {
155                break;
156            }
157        }
158        return new int[] { matches, transpositions / 2, prefix, max.length() };
159    }
160
161}