public void TestDistanceCharacterInsDelInterface()
{
var instance = new WeightedLevenshtein(new ExampleCharSub(), new ExampleInsDel());
// Same as testDistance above.
Assert.Equal(0.0, instance.Distance("String1", "String1"), 1);
Assert.Equal(0.5, instance.Distance("String1", "Srring1"), 1);
Assert.Equal(1.5, instance.Distance("String1", "Srring2"), 1);
// Cost of insert of 'i' is less than normal, so these scores are
// different than testDistance above. Note that the cost of delete
// has been set differently than the cost of insert, so the distance
// call is not symmetric in its arguments if an 'i' has changed.
Assert.Equal(0.5, instance.Distance("Strng", "String"), 1);
Assert.Equal(0.8, instance.Distance("String", "Strng"), 1);
Assert.Equal(1.0, instance.Distance("Strig", "String"), 1);
Assert.Equal(1.0, instance.Distance("String", "Strig"), 1);
// Same as above with limits.
Assert.Equal(0.0, instance.Distance("String1", "String1", double.MaxValue), 1);
Assert.Equal(0.0, instance.Distance("String1", "String1", 2.0), 1);
Assert.Equal(1.5, instance.Distance("String1", "Srring2", double.MaxValue), 1);
Assert.Equal(1.5, instance.Distance("String1", "Srring2", 2.0), 1);
Assert.Equal(1.5, instance.Distance("String1", "Srring2", 1.5), 1);
Assert.Equal(1.0, instance.Distance("String1", "Srring2", 1.0), 1);
Assert.Equal(4.0, instance.Distance("String1", "Potato", 4.0), 1);
NullEmptyTests.TestDistance(instance);
}
public void TestDistance()
{
var instance = new QGram(k: 2);
// AB BC CD CE
// 1 1 1 0
// 1 1 0 1
// Total: 2
var result = instance.Distance("ABCD", "ABCE");
Assert.Equal(expected: 2.0, actual: result);
Assert.Equal(
expected: 0.0,
actual: instance.Distance("S", "S"),
precision: 1); // 0.0
Assert.Equal(
expected: 0.0,
actual: instance.Distance("012345", "012345"),
precision: 1); // 0.0
// NOTE: not using null/empty tests in NullEmptyTests because QGram is different
Assert.Equal(0.0, instance.Distance("", ""), 1);
Assert.Equal(2.0, instance.Distance("", "foo"), 1);
Assert.Equal(2.0, instance.Distance("foo", ""), 1);
NullEmptyTests.AssertArgumentNullExceptions(instance);
}
public void TestDistance()
{
var s0 = "ABABABAB";
var s1 = "ABCABCABCABC";
var s2 = "POIULKJH";
var ngram = new NGram();
Assert.True(ngram.Distance(s0, s1) < ngram.Distance(s0, s2));
Assert.Equal(
expected: 0.0,
actual: ngram.Distance("SIJK", "SIJK"),
precision: 1); // 0.0
Assert.Equal(
expected: 0.0,
actual: ngram.Distance("S", "S"),
precision: 1); // 0.0
Assert.Equal(
expected: 1.0,
actual: ngram.Distance("", "S"),
precision: 1); // 0.0
Assert.Equal(
expected: 1.0,
actual: ngram.Distance("", "SIJK"),
precision: 1); // 0.0
NullEmptyTests.TestDistance(ngram);
}
public void TestDistance()
{
var instance = new MetricLCS();
NullEmptyTests.TestDistance(instance);
// TODO: regular (non-null/empty) distance tests
}
public void TestDistance()
{
var instance = new JaroWinkler();
NullEmptyTests.TestDistance(instance);
// TODO: regular (non-null/empty) distance tests
}
public void TestSimilarity()
{
var instance = new NormalizedLevenshtein();
NullEmptyTests.TestSimilarity(instance);
// TODO: regular (non-null/empty) similarity tests
}
public void TestDistance()
{
var instance = new RatcliffObershelp();
NullEmptyTests.TestDistance(instance);
/// TODO: regular (non-null/empty) distance tests
}
public void TestDistance()
{
var instance = new NormalizedLevenshtein();
NullEmptyTests.TestDistance(instance);
// TODO: regular (non-null/empty) distance tests
}
public void TestDistance()
{
var instance = new WeightedLevenshtein(new ExampleCharSub());
Assert.Equal(0.0, instance.Distance("String1", "String1"), 1);
Assert.Equal(0.5, instance.Distance("String1", "Srring1"), 1);
Assert.Equal(1.5, instance.Distance("String1", "Srring2"), 1);
NullEmptyTests.TestDistance(instance);
}
public void TestDistance()
{
var instance = new Damerau();
Assert.Equal(expected: 1.0, actual: instance.Distance("ABCDEF", "ABDCEF"));
Assert.Equal(expected: 2.0, actual: instance.Distance("ABCDEF", "BACDFE"));
Assert.Equal(expected: 1.0, actual: instance.Distance("ABCDEF", "ABCDE"));
NullEmptyTests.TestDistance(instance);
}
public void TestDistance()
{
var instance = new Levenshtein();
Assert.Equal(expected: 1.0, actual: instance.Distance("My string", "My tring"));
Assert.Equal(expected: 2.0, actual: instance.Distance("My string", "M string2"));
Assert.Equal(expected: 1.0, actual: instance.Distance("My string", "My $tring"));
NullEmptyTests.TestDistance(instance);
}
public void TestDistance()
{
var instance = new Jaccard(k: 2);
var result = instance.Distance("ABCDE", "ABCDF");
Assert.Equal(expected: 0.4, actual: result);
NullEmptyTests.TestDistance(instance);
}
public void TestDistance()
{
var instance = new LongestCommonSubsequence();
// LCS = GA or GC => distance = 4 (remove 3 letters and add 1)
Assert.Equal(expected: 4, actual: instance.Distance("AGCAT", "GAC"));
Assert.Equal(expected: 1, actual: instance.Distance("AGCAT", "AGCT"));
NullEmptyTests.TestDistance(instance);
}
public void TestSimilarity()
{
var instance = new Cosine();
var result = instance.Similarity("ABC", "ABCE");
Assert.Equal(
expected: 0.71,
actual: result,
precision: 2 // 0.01
);
NullEmptyTests.TestSimilarity(instance);
}
public void TestSimilarity()
{
var instance = new Jaccard(k: 2);
// AB BC CD DE DF
// 1 1 1 1 0
// 1 1 1 0 1
// => 3 / 5 = 0.6
var result = instance.Similarity("ABCDE", "ABCDF");
Assert.Equal(expected: 0.6, actual: result);
NullEmptyTests.TestSimilarity(instance);
}
public void TestSimilarity()
{
var instance = new SorensenDice(2);
// AB BC CD DE DF FG
// 1 1 1 1 0 0
// 1 1 1 0 1 1
// => 2 x 3 / (4 + 5) = 6/9 = 0.6666
var result = instance.Similarity("ABCDE", "ABCDFG");
Assert.Equal(
expected: 0.6667, // last digit should be 7 because of rounding
actual: result,
precision: 4 //0.0001
);
NullEmptyTests.TestSimilarity(instance);
}
public void TestSimilarity()
{
var instance = new JaroWinkler();
Assert.Equal(
expected: 0.974074,
actual: instance.Similarity("My string", "My tsring"),
precision: 6 // 0.000001
);
Assert.Equal(
expected: 0.896296,
actual: instance.Similarity("My string", "My ntrisg"),
precision: 6 // 0.000001
);
NullEmptyTests.TestSimilarity(instance);
}
public void TestDistance()
{
var instance = new WeightedLevenshtein(new ExampleCharSub());
Assert.Equal(0.0, instance.Distance("String1", "String1"), 1);
Assert.Equal(0.5, instance.Distance("String1", "Srring1"), 1);
Assert.Equal(1.5, instance.Distance("String1", "Srring2"), 1);
// One insert or delete.
Assert.Equal(1.0, instance.Distance("Strng", "String"), 1);
Assert.Equal(1.0, instance.Distance("String", "Strng"), 1);
// With limits.
Assert.Equal(0.0, instance.Distance("String1", "String1", double.MaxValue), 1);
Assert.Equal(0.0, instance.Distance("String1", "String1", 2.0), 1);
Assert.Equal(1.5, instance.Distance("String1", "Srring2", double.MaxValue), 1);
Assert.Equal(1.5, instance.Distance("String1", "Srring2", 2.0), 1);
Assert.Equal(1.5, instance.Distance("String1", "Srring2", 1.5), 1);
Assert.Equal(1.0, instance.Distance("String1", "Srring2", 1.0), 1);
Assert.Equal(4.0, instance.Distance("String1", "Potato", 4.0), 1);
NullEmptyTests.TestDistance(instance);
}
public void TestDistance()
{
var instance = new OptimalStringAlignment();
// zero length
Assert.Equal(
expected: 6.0,
actual: instance.Distance("", "ABCDEF"),
precision: 0 // 0.0
);
Assert.Equal(
expected: 6.0,
actual: instance.Distance("ABCDEF", ""),
precision: 0 // 0.0
);
Assert.Equal(
expected: 0.0,
actual: instance.Distance("", ""),
precision: 0 // 0.0
);
// equality
Assert.Equal(
expected: 0.0,
actual: instance.Distance("ABCDEF", "ABCDEF"),
precision: 0 // 0.0
);
// single operation
Assert.Equal(
expected: 1.0,
actual: instance.Distance("ABDCFE", "ABDCEF"),
precision: 0 // 0.0
);
Assert.Equal(
expected: 1.0,
actual: instance.Distance("BBDCEF", "ABDCEF"),
precision: 0 // 0.0
);
Assert.Equal(
expected: 1.0,
actual: instance.Distance("BDCEF", "ABDCEF"),
precision: 0 // 0.0
);
Assert.Equal(
expected: 1.0,
actual: instance.Distance("ABDCEF", "ADCEF"),
precision: 0 // 0.0
);
// other
Assert.Equal(
expected: 3.0,
actual: instance.Distance("CA", "ABC"),
precision: 0 // 0.0
);
Assert.Equal(
expected: 2.0,
actual: instance.Distance("BAC", "CAB"),
precision: 0 // 0.0
);
Assert.Equal(
expected: 4.0,
actual: instance.Distance("abcde", "awxyz"),
precision: 0 // 0.0
);
Assert.Equal(
expected: 5.0,
actual: instance.Distance("abcde", "vwxyz"),
precision: 0 // 0.0
);
NullEmptyTests.TestDistance(instance);
}
public void TestSimilarity()
{
var instance = new RatcliffObershelp();
/// test data from other algorithms
/// "My string" vs "My tsring"
/// Substrings:
/// "ring" ==> 4, "My s" ==> 3, "s" ==> 1
/// Ratcliff-Obershelp = 2*(sum of substrings)/(length of s1 + length of s2)
/// = 2*(4 + 3 + 1) / (9 + 9)
/// = 16/18
/// = 0.888888888888889
Assert.Equal(
expected: 0.888888888888889,
actual: instance.Similarity("My string", "My tsring"),
precision: 15 /// 0.000000000000001
);
/// test data from other algorithms
/// "My string" vs "My tsring"
/// Substrings:
/// "My " ==> 3, "tri" ==> 3, "g" ==> 1
/// Ratcliff-Obershelp = 2*(sum of substrings)/(length of s1 + length of s2)
/// = 2*(3 + 3 + 1) / (9 + 9)
/// = 14/18
/// = 0.777777777777778
Assert.Equal(
expected: 0.777777777777778,
actual: instance.Similarity("My string", "My ntrisg"),
precision: 15 /// 0.000000000000001
);
/// test data from essay by Ilya Ilyankou
/// "Comparison of Jaro-Winkler and Ratcliff/Obershelp algorithms
/// in spell check"
/// https://ilyankou.files.wordpress.com/2015/06/ib-extended-essay.pdf
/// p13, expected result is 0.857
Assert.Equal(
expected: 0.857,
actual: instance.Similarity("MATEMATICA", "MATHEMATICS"),
precision: 3 /// 0.001
);
/// test data from stringmetric
/// https://github.com/rockymadden/stringmetric
/// expected output is 0.736842105263158
Assert.Equal(
expected: 0.736842105263158,
actual: instance.Similarity("aleksander", "alexandre"),
precision: 15 /// 0.000000000000001
);
/// test data from stringmetric
/// https://github.com/rockymadden/stringmetric
/// expected output is 0.666666666666667
Assert.Equal(
expected: 0.666666666666667,
actual: instance.Similarity("pennsylvania", "pencilvaneya"),
precision: 15 /// 0.000000000000001
);
/// test data from wikipedia
/// https://en.wikipedia.org/wiki/Gestalt_Pattern_Matching
/// expected output is 14/18 = 0.777777777777778
Assert.Equal(
expected: 0.777777777777778,
actual: instance.Similarity("WIKIMEDIA", "WIKIMANIA"),
precision: 15 /// 0.000000000000001
);
/// test data from wikipedia
/// https://en.wikipedia.org/wiki/Gestalt_Pattern_Matching
/// expected output is 24/40 = 0.6
Assert.Equal(
expected: 0.6,
actual: instance.Similarity("GESTALT PATTERN MATCHING", "GESTALT PRACTICE"),
precision: 15 /// 0.000000000000001
);
NullEmptyTests.TestSimilarity(instance);
}
public void TestDistance()
{
var instance = new SorensenDice();
NullEmptyTests.TestDistance(instance);
}
