Rosetta Code

Sorensen–Dice coefficient: Difference between revisions

Page Discussion

Sorensen–Dice coefficient (view source)

Revision as of 20:04, 10 March 2024

21,755 bytes added ,  2 months ago
m
→‎{{header|jq}}: simplify
m (→‎{{header|Phix}}: missed "Bell numbers", extended comment)
m (→‎{{header|jq}}: simplify)
 
(33 intermediate revisions by 10 users not shown)
Line 1:
{{draft task}}
 
The [[wp:Sørensen–Dice coefficient|Sørensen–Dice coefficient]], also known as the Sørensen–Dice index (or sdi, or sometimes by one of the individual names: sorensen or dice,) is a statistic used to gauge the similarity of two poulationpopulation samples.
 
The original use was in botany, indexingas a measure of similarity between populations of flora and fauna in different areas, but it has uses in other fields as well. It can be used as a text similarity function somewhat similar to the [[Levenshtein distance|Levenshtein edit distance]] function, though it'sits strengthcharacteristics liesare in aquite different area.
 
[[Levenshtein distance|Levenshtein]] iscan be gooduseful for findingspelling misspellingscorrection, but relies on the tested word /or phrase being prettyquite similar to the desired one, and can be very slow for long words /or phrases.
 
Sørensen–Dice is more useful for 'fuzzy' matching partial, and poorly spelled words /or phrases, possibly in improper order.
 
There are several different methods to tokenize objects for Sørensen–Dice comparisons. The most typical tokenizing scheme for text is to break the words up into bi-grams: groups of two consecutive letters.
Line 22:
 
Sørensen–Dice measures the similarity of two groups by dividing twice the intersection token count by the total token count of both groups.:
 
SDC = 2 × |A∩B| / (|A| + |B|)
For items(objects, populations) A and B:
 
where A, B and A∩B are to be understood as multisets, and that if an item, x, has multiplicity a in A and b in B, then it will have multiplicity min(a,b) in A∩B.
SDI = 2 × (A ∩ B) / (A ⊎ B)
 
The Sørensen–Dice coefficient is thus a ratio between 0.0 and 1.0 giving the "percent similarity" between the two populations between 0.0 and 1.0.
 
SDISDC ''can'' by used for spellchecking, but it's not really good at it, especially for short words. Where it really shines is for fuzzy matching of short phrases like book or movie titles. It may not return exactly what you are looking for, but often gets remarkably close with some pretty poor inputs.
 
 
Line 41:
How you get the task names is peripheral to the task. You can [[:Category:Programming_Tasks|web-scrape]] them or [[Sorensen–Dice coefficient/Tasks|download them to a file]], whatever.
 
If there is a built-in or easily, freely available library implementation for Sørensen–Dice coefficient calculations, it is acceptable to use that with a pointer to where it may be obtained.
 
 
=={{header|C++}}==
{{trans|Wren}}
<syntaxhighlight lang="cpp">#include <algorithm>
#include <cctype>
#include <cstdlib>
#include <fstream>
#include <iostream>
#include <iterator>
#include <set>
#include <sstream>
#include <string>
#include <vector>
 
using bigram = std::pair<char, char>;
 
std::multiset<bigram> bigrams(const std::string& phrase) {
std::multiset<bigram> result;
std::istringstream is(phrase);
std::string word;
while (is >> word) {
for (char& ch : word) {
ch = std::tolower(static_cast<unsigned char>(ch));
}
size_t length = word.size();
if (length == 1) {
result.emplace(word[0], '\0');
} else {
for (size_t i = 0; i + 1 < length; ++i) {
result.emplace(word[i], word[i + 1]);
}
}
}
return result;
}
 
double sorensen(const std::string& s1, const std::string& s2) {
auto a = bigrams(s1);
auto b = bigrams(s2);
std::multiset<bigram> c;
std::set_intersection(a.begin(), a.end(), b.begin(), b.end(),
std::inserter(c, c.begin()));
return (2.0 * c.size()) / (a.size() + b.size());
}
 
int main() {
std::vector<std::string> tasks;
std::ifstream is("tasks.txt");
if (!is) {
std::cerr << "Cannot open tasks file.\n";
return EXIT_FAILURE;
}
std::string task;
while (getline(is, task)) {
tasks.push_back(task);
}
const size_t tc = tasks.size();
const std::string tests[] = {"Primordial primes",
"Sunkist-Giuliani formula",
"Sieve of Euripides", "Chowder numbers"};
std::vector<std::pair<double, size_t>> sdi(tc);
std::cout << std::fixed;
for (const std::string& test : tests) {
for (size_t i = 0; i != tc; ++i) {
sdi[i] = std::make_pair(sorensen(tasks[i], test), i);
}
std::partial_sort(sdi.begin(), sdi.begin() + 5, sdi.end(),
[](const std::pair<double, size_t>& a,
const std::pair<double, size_t>& b) {
return a.first > b.first;
});
std::cout << test << " >\n";
for (size_t i = 0; i < 5 && i < tc; ++i) {
std::cout << " " << sdi[i].first << ' ' << tasks[sdi[i].second]
<< '\n';
}
std::cout << '\n';
}
return EXIT_SUCCESS;
}</syntaxhighlight>
 
{{out}}
<pre>
Primordial primes >
0.685714 Sequence of primorial primes
0.666667 Factorial primes
0.571429 Primorial numbers
0.545455 Prime words
0.521739 Almost prime
 
Sunkist-Giuliani formula >
0.565217 Almkvist-Giullera formula for pi
0.378378 Faulhaber's formula
0.342857 Haversine formula
0.333333 Check Machin-like formulas
0.307692 Resistance calculator
 
Sieve of Euripides >
0.461538 Four sides of square
0.461538 Sieve of Pritchard
0.413793 Sieve of Eratosthenes
0.400000 Piprimes
0.384615 Sierpinski curve
 
Chowder numbers >
0.782609 Chowla numbers
0.640000 Powerful numbers
0.608696 Rhonda numbers
0.608696 Fermat numbers
0.600000 Lah numbers
 
</pre>
 
=={{header|J}}==
Tentative implementation:
<syntaxhighlight lang=J>TASKS=: fread '~/tasks.txt' NB. from Sorensen–Dice_coefficient/Tasks
sdtok=: [: (#~ ' '*/ .~:~])2]\ 7 u: tolower@rplc&(LF,' ')
sdinter=: {{
all=. ~.x,y
X=. <:#/.~all,x
Y=. <:#/.~all,y
+/X<.Y
}}
sdunion=: #@,
SDC=: (2 * sdinter % sdunion)&sdtok S:0
nearest=: {{ m{.\:~ x (] ;"0~ SDC) cutLF y }}
fmt=: ((8j6": 0{::]),' ',1{::])"1</syntaxhighlight>
 
The trick here is the concept of "intersection" which we must use. We can't use set intersection -- the current draft task description suggests that <code>SDI = 2 × (A ∩ B) / (A ⊎ B)</code> produces a number between 0 and 1. Because we're using division to produce this number, we must be using cardinality of the intersection rather than the intersection itself.
But if A and B are sets, each containing the same tokens, the result here using cardinality of sets would be 2 rather than 1.
 
Instead, we treat A and B as sequences of tokens (so repeated copies of a token are distinct), for the cardinality of the intersection we count the number of times that each token appears in either A and in B and sum the minimum of the two counts. (So, tokens which only appear in A count 0 times, for example, where a token which appears 3 times in A and 2 times in B would contribute 2 to the sum.)
 
With this implementation, here's the task examples:
<pre> fmt 'Primordial prime' 5 nearest TASKS
0.647059 Sequence of primorial primes
0.615385 Factorial primes
0.592593 Primorial numbers
0.571429 Prime words
0.545455 Almost prime
fmt 'Sunkist-Giuliani formula' 5 nearest TASKS
0.565217 Almkvist-Giullera formula for pi
0.378378 Faulhaber's formula
0.342857 Haversine formula
0.333333 Check Machin-like formulas
0.307692 Resistance calculator
fmt 'Sieve of Euripides' 5 nearest TASKS
0.461538 Sieve of Pritchard
0.461538 Four sides of square
0.413793 Sieve of Eratosthenes
0.400000 Piprimes
0.384615 Sierpinski curve
fmt 'Chowder numbers' 5 nearest TASKS
0.782609 Chowla numbers
0.640000 Powerful numbers
0.608696 Rhonda numbers
0.608696 Fermat numbers
0.600000 Lah numbers </pre>
 
=={{header|Java}}==
<syntaxhighlight lang="java">
import java.io.IOException;
import java.nio.charset.StandardCharsets;
import java.nio.file.Files;
import java.nio.file.Path;
import java.util.ArrayList;
import java.util.Collections;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
 
public final class SorensenDiceCoefficient {
 
public static void main(String[] args) throws IOException {
List<String> tasks = Files.readAllLines(Path.of("Rosetta Code Tasks.dat"), StandardCharsets.UTF_8);
List<String> tests = List.of(
"Primordial primes", "Sunkist-Giuliani formula", "Sieve of Euripides", "Chowder numbers" );
record TaskValue(String task, double value) {}
for ( String test : tests ) {
List<TaskValue> taskValues = new ArrayList<TaskValue>();
Map<String, Integer> bigramsTest = createBigrams(test);
for ( String task : tasks ) {
double value = sorensenDice(bigramsTest, createBigrams(task));
taskValues.add( new TaskValue(task, value) );
}
Collections.sort(taskValues, (one, two) -> Double.compare(two.value, one.value));
System.out.println(test + ":");
for ( int i = 0; i < 5; i++ ) {
System.out.println(String.format("%s%.4f%s%s",
" ", taskValues.get(i).value, " ", taskValues.get(i).task));
}
System.out.println();
}
}
private static double sorensenDice(Map<String, Integer> bigramsOne, Map<String, Integer> bigramsTwo) {
int intersectionSize = 0;
for ( Map.Entry<String, Integer> entry : bigramsOne.entrySet() ) {
if ( bigramsTwo.keySet().contains(entry.getKey()) ) {
intersectionSize += Math.min(entry.getValue(), bigramsTwo.get(entry.getKey()));
}
}
return 2.0 * intersectionSize / ( size(bigramsOne) + size(bigramsTwo) );
}
private static Map<String, Integer> createBigrams(String text) {
Map<String, Integer> result = new HashMap<String, Integer>();
for ( String word : text.toLowerCase().split(" ") ) {
if ( word.length() == 1 ) {
result.merge(word, 1, Integer::sum);
} else {
for ( int i = 0; i < word.length() - 1; i++ ) {
result.merge(word.substring(i, i + 2), 1, Integer::sum);
}
}
}
return result;
}
private static int size(Map<String, Integer> map) {
return map.values().stream().mapToInt(Integer::intValue).sum();
}
 
}
</syntaxhighlight>
{{ out }}
<pre>
Primordial primes:
0.6857 Sequence of primorial primes
0.6667 Factorial primes
0.5714 Primorial numbers
0.5455 Prime words
0.5217 Almost prime
 
Sunkist-Giuliani formula:
0.5652 Almkvist-Giullera formula for pi
0.3784 Faulhaber's formula
0.3429 Haversine formula
0.3333 Check Machin-like formulas
0.3077 Resistance calculator
 
Sieve of Euripides:
0.4615 Four sides of square
0.4615 Sieve of Pritchard
0.4138 Sieve of Eratosthenes
0.4000 Piprimes
0.3846 Sierpinski curve
 
Chowder numbers:
0.7826 Chowla numbers
0.6400 Powerful numbers
0.6087 Fermat numbers
0.6087 Rhonda numbers
0.6000 Lah numbers
</pre>
 
=={{header|jq}}==
{{Works with|jq}}
 
'''Works with gojq, the Go implementation of jq'''
 
'''Works with jaq, the Rust implementation of jq'''
 
'''Adapted from [[#Wren|Wren]]'''
<syntaxhighlight lang="jq">
### Generic preliminaries
 
def count(s): reduce s as $x (0; .+1);
 
def lpad($len): tostring | ($len - length) as $l | (" " * $l) + .;
 
# Emit the count of the common items in the two given sorted arrays
# viewed as multisets
def count_commonality_of_multisets($A; $B):
# Returns a stream of the common elements
def pop:
.[0] as $i
| .[1] as $j
| if $i == ($A|length) or $j == ($B|length) then empty
elif $A[$i] == $B[$j] then 1, ([$i+1, $j+1] | pop)
elif $A[$i] < $B[$j] then [$i+1, $j] | pop
else [$i, $j+1] | pop
end;
count([0,0] | pop);
 
# Emit an array of the normalized bigrams of the input string
def bigrams:
# Emit a stream of the bigrams of the input string blindly
def bg: . as $in | range(0;length-1 ) | $in[.:.+2];
ascii_downcase | [splits(" *") | bg];
 
 
### The Sorensen-Dice coefficient
 
def sorensen($a; $b):
($a | bigrams | sort) as $A
| ($b | bigrams | sort) as $B
| 2 * count_commonality_of_multisets($A; $B) / (($A|length) + ($B|length));
 
 
### Exercises
 
def exercises:
"Primordial primes",
"Sunkist-Giuliani formula",
"Sieve of Euripides",
"Chowder numbers"
;
 
[inputs] as $phrases
| exercises as $test
| [ range(0; $phrases|length) as $i
| [sorensen($phrases[$i]; $test), $phrases[$i] ] ]
| sort_by(first)
| .[-5:]
| reverse
| "\($test) >",
map( " \(first|tostring|.[:4]|lpad(4)) \(.[1])")[],
""
</syntaxhighlight>
{{output}}
Invocation: jq -nrR -f sorensen-dice-coefficient.jq rc_tasks_2022_09_24.txt
<pre>
Primordial primes >
0.68 Sequence of primorial primes
0.66 Factorial primes
0.57 Primorial numbers
0.54 Prime words
0.52 Almost prime
 
Sunkist-Giuliani formula >
0.56 Almkvist-Giullera formula for pi
0.37 Faulhaber's formula
0.34 Haversine formula
0.33 Check Machin-like formulas
0.30 Resistance calculator
 
Sieve of Euripides >
0.46 Sieve of Pritchard
0.46 Four sides of square
0.41 Sieve of Eratosthenes
0.4 Piprimes
0.38 Sierpinski curve
 
Chowder numbers >
0.78 Chowla numbers
0.64 Powerful numbers
0.60 Rhonda numbers
0.60 Fermat numbers
0.6 Lah numbers
</pre>
 
=={{header|Julia}}==
 
<syntaxhighlight lang="julia">using Multisets
 
""" convert a phrase into a count of bigram tokens of its words """
function tokenizetext(txt)
tokens = Multiset{String}()
words = split(lowercase(txt), r"\s+")
for w in words
a = collect(w)
if length(a) < 3
push!(tokens, w)
else
for i in 1:length(a)-1
push!(tokens, String(a[i:i+1]))
end
end
end
return tokens
end
 
""" Sorenson-Dice similarity of multisets """
function sorenson_dice(text1, text2)
bc1, bc2 = tokenizetext(text1), tokenizetext(text2)
return 2 * length(bc1 ∩ bc2) / (length(bc1) + length(bc2))
end
 
const alltasks = split(read("onedrive/documents/julia programs/tasks.txt", String), "\n")
 
# run tests
for test in ["Primordial primes", "Sunkist-Giuliani formula",
"Sieve of Euripides", "Chowder numbers"]
taskvalues = sort!([(sorenson_dice(test, t), t) for t in alltasks], rev = true)
println("\n$test:")
for (val, task) in taskvalues[begin:begin+4]
println(lpad(Float16(val), 8), " ", task)
end
end
 
</syntaxhighlight>{{out}}
<pre>
Primordial primes:
0.6855 Sequence of primorial primes
0.6665 Factorial primes
0.5713 Primorial numbers
0.5454 Prime words
0.522 Almost prime
 
Sunkist-Giuliani formula:
0.5654 Almkvist-Giullera formula for pi
0.3784 Faulhaber's formula
0.3428 Haversine formula
0.3333 Check Machin-like formulas
0.3076 Resistance calculator
 
Sieve of Euripides:
0.4614 Sieve of Pritchard
0.4614 Four sides of square
0.4138 Sieve of Eratosthenes
0.4 Piprimes
0.3845 Sierpinski curve
 
Chowder numbers:
0.7827 Chowla numbers
0.64 Powerful numbers
0.609 Rhonda numbers
0.609 Fermat numbers
0.6 Lah numbers
</pre>
 
=={{header|Nim}}==
<syntaxhighlight lang=Nim>import std/[algorithm, strutils, sugar, tables]
 
func bigrams(text: string): CountTable[string] =
## Extract the bigrams from a text.
for word in text.toLower.split(' '):
if word.len == 1:
result.inc(word)
else:
for i in 0..(word.len - 2):
result.inc(word[i..(i+1)])
 
func intersectionCount(a, b: CountTable[string]): int =
## Compute the cardinal of the intersection of two
## count tables.
for key, count in a:
if key in b:
inc result, min(count, b[key])
 
func card(a: CountTable[string]): int =
## Return the cardinal of a count table (i.e. the sum of counts).
for count in a.values:
inc result, count
 
func sorensenDice(text1, text2: string): float =
## Compute the Sorensen-dice coefficient of "text1" and "text2".
let ct1 = text1.bigrams
let ct2 = text2.bigrams
result = 2 * intersectionCount(ct1, ct2) / (ct1.card + ct2.card)
 
# Build the list of tasks.
let tasks = collect:
for line in lines("Sorensen-Dice.txt"):
line
 
const Tests = ["Primordial primes", "Sunkist-Giuliani formula",
"Sieve of Euripides", "Chowder numbers"]
 
for test in Tests:
echo test
var scores: seq[(float, string)]
for task in tasks:
scores.add (sorensenDice(test, task), task)
scores.sort(Descending)
for i in 0..4:
echo " ", scores[i][0].formatFloat(ffDecimal, 6), ' ', scores[i][1]
echo()
</syntaxhighlight>
 
{{out}}
<pre>Primordial primes
0.685714 Sequence of primorial primes
0.666667 Factorial primes
0.571429 Primorial numbers
0.545455 Prime words
0.521739 Almost prime
 
Sunkist-Giuliani formula
0.565217 Almkvist-Giullera formula for pi
0.378378 Faulhaber's formula
0.342857 Haversine formula
0.333333 Check Machin-like formulas
0.307692 Resistance calculator
 
Sieve of Euripides
0.461538 Sieve of Pritchard
0.461538 Four sides of square
0.413793 Sieve of Eratosthenes
0.400000 Piprimes
0.384615 Sierpinski curve
 
Chowder numbers
0.782609 Chowla numbers
0.640000 Powerful numbers
0.608696 Rhonda numbers
0.608696 Fermat numbers
0.600000 Lah numbers
</pre>
 
=={{header|Perl}}==
<syntaxhighlight lang="perl" line>use v5.036;
use Path::Tiny;
use List::Util <uniq head>;
 
sub bi_gram {
my $line = lc shift;
uniq map { substr $line,$_,2 } 0..length($line)-2;
}
 
sub score {
my($phrase, $word) = @_;
my %count;
my @match = bi_gram $phrase;
$count{$_}++ for @match, @$word;
2 * (grep { $count{$_} > 1 } keys %count) / (@match + @$word);
}
 
sub sorensen {
my($dict,$word,$cutoff) = @_; $cutoff //= 0.00;
my(%matches,$s);
($s = score($word, $$dict{$_})) > $cutoff and $matches{$_} = $s for keys %$dict;
%matches;
}
 
my %dict = map { $_ => [ bi_gram($_) ] } path('ref/Sorensen-Dice-Tasks.txt')->slurp =~ /.{10,}/gm;
 
for my $word ( ('Primordial primes', 'Sunkist-Giuliani formula', 'Sieve of Euripides', 'Chowder numbers') ) {
my(%scored,@ranked);
%scored = sorensen(\%dict,$word);
push @ranked, sprintf "%.3f $_", $scored{$_} for sort { $scored{$b} <=> $scored{$a} || $a cmp $b } keys %scored;
say "\n$word:\n" . join("\n", head 5, @ranked);
}</syntaxhighlight>
{{out}}
<pre>Primordial primes:
0.741 Factorial primes
0.629 Sequence of primorial primes
0.583 Almost prime
0.581 Next special primes
0.571 Pandigital prime
 
Sunkist-Giuliani formula:
0.542 Almkvist-Giullera formula for pi
0.368 Haversine formula
0.359 Faulhaber's formula
0.348 Check Machin-like formulas
0.303 FASTA format
 
Sieve of Euripides:
0.541 Sieve of Eratosthenes
0.529 Sieve of Pritchard
0.457 Four sides of square
0.457 The sieve of Sundaram
0.387 Sum of a series
 
Chowder numbers:
0.769 Chowla numbers
0.615 Rhonda numbers
0.609 Bell numbers
0.609 Lah numbers
0.593 Kaprekar numbers</pre>
 
=={{header|Phix}}==
<!--<syntaxhighlight lang="phix">(phixonline)-->
<span style="color: #008080;">with</span> <span style="color: #008080;">javascript_semantics</span>
<span style="color: #008080;">constant</span> <span style="color: #000000;">match_raku</span> <span style="color: #0000FF;">=</span> <span style="color: #004600;">false</span>
<span style="color: #008080;">include</span> <span style="color: #000000;">sets</span><span style="color: #0000FF;">.</span><span style="color: #000000;">e</span>
<span style="color: #008080;">function</span> <span style="color: #000000;">bigram</span><span style="color: #0000FF;">(</span><span style="color: #004080;">string</span> <span style="color: #000000;">s</span><span style="color: #0000FF;">)</span>
<span style="color: #004080;">sequence</span> <span style="color: #000000;">words</span> <span style="color: #0000FF;">=</span> <span style="color: #7060A8;">split</span><span style="color: #0000FF;">(</span><span style="color: #7060A8;">lower</span><span style="color: #0000FF;">(</span><span style="color: #000000;">s</span><span style="color: #0000FF;">)),</span>
Line 55 ⟶ 619:
<span style="color: #008080;">for</span> <span style="color: #000000;">word</span> <span style="color: #008080;">in</span> <span style="color: #000000;">words</span> <span style="color: #008080;">do</span>
<span style="color: #008080;">for</span> <span style="color: #000000;">i</span><span style="color: #0000FF;">=</span><span style="color: #000000;">1</span> <span style="color: #008080;">to</span> <span style="color: #7060A8;">length</span><span style="color: #0000FF;">(</span><span style="color: #000000;">word</span><span style="color: #0000FF;">)-</span><span style="color: #000000;">1</span> <span style="color: #008080;">do</span>
<span style="color: #008080000000;">ifres</span> <span style="color: #0080800000FF;">not=</span> <span style="color: #7060A8;">append</span><span style="color: #0000FF;">(</span><span style="color: #000000;">match_rakures</span><span style="color: #0000FF;">,</span><span style="color: #000000;">word</span><span style="color: #0000FF;">[</span><span style="color: #000000;">i</span><span style="color: #0000FF;">..</span><span style="color: #000000;">i</span><span style="color: #0000FF;">+</span><span style="color: #000000;">1</span><span style="color: #0080800000FF;">then])</span>
<span style="color: #000000;">res</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">add_member</span><span style="color: #0000FF;">(</span><span style="color: #000000;">res</span><span style="color: #0000FF;">,</span><span style="color: #000000;">word</span><span style="color: #0000FF;">[</span><span style="color: #000000;">i</span><span style="color: #0000FF;">..</span><span style="color: #000000;">i</span><span style="color: #0000FF;">+</span><span style="color: #000000;">1</span><span style="color: #0000FF;">])</span>
<span style="color: #008080;">else</span>
<span style="color: #004080;">string</span> <span style="color: #000000;">c2</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">word</span><span style="color: #0000FF;">[</span><span style="color: #000000;">i</span><span style="color: #0000FF;">..</span><span style="color: #000000;">i</span><span style="color: #0000FF;">+</span><span style="color: #000000;">1</span><span style="color: #0000FF;">]</span>
<span style="color: #008080;">while</span> <span style="color: #000000;">is_member</span><span style="color: #0000FF;">(</span><span style="color: #000000;">res</span><span style="color: #0000FF;">,</span><span style="color: #000000;">c2</span><span style="color: #0000FF;">)</span> <span style="color: #008080;">do</span>
<span style="color: #000000;">c2</span> <span style="color: #0000FF;">&=</span> <span style="color: #008000;">'1'</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">while</span>
<span style="color: #000000;">res</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">add_member</span><span style="color: #0000FF;">(</span><span style="color: #000000;">res</span><span style="color: #0000FF;">,</span><span style="color: #000000;">c2</span><span style="color: #0000FF;">)</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">if</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">for</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">for</span>
<span style="color: #000000;">res</span> <span style="color: #0000FF;">=</span> <span style="color: #7060A8;">sort</span><span style="color: #0000FF;">(</span><span style="color: #000000;">res</span><span style="color: #0000FF;">)</span>
<span style="color: #008080;">return</span> <span style="color: #000000;">res</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">function</span>
<span style="color: #008080;">function</span> <span style="color: #000000;">intrasect</span><span style="color: #0000FF;">(</span><span style="color: #004080;">sequence</span> <span style="color: #000000;">s1</span><span style="color: #0000FF;">,</span> <span style="color: #000000;">s2</span><span style="color: #0000FF;">)</span>
<span style="color: #004080;">integer</span> <span style="color: #000000;">l1</span> <span style="color: #0000FF;">=</span> <span style="color: #7060A8;">length</span><span style="color: #0000FF;">(</span><span style="color: #000000;">s1</span><span style="color: #0000FF;">),</span>
<span style="color: #000000;">l2</span> <span style="color: #0000FF;">=</span> <span style="color: #7060A8;">length</span><span style="color: #0000FF;">(</span><span style="color: #000000;">s2</span><span style="color: #0000FF;">),</span>
<span style="color: #000000;">i1</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">1</span><span style="color: #0000FF;">,</span> <span style="color: #000000;">i2</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">1</span><span style="color: #0000FF;">,</span>
<span style="color: #000000;">res</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">0</span>
<span style="color: #008080;">while</span> <span style="color: #000000;">i1</span><span style="color: #0000FF;"><=</span><span style="color: #000000;">l1</span> <span style="color: #008080;">and</span> <span style="color: #000000;">i2</span><span style="color: #0000FF;"><=</span><span style="color: #000000;">l2</span> <span style="color: #008080;">do</span>
<span style="color: #004080;">integer</span> <span style="color: #000000;">c</span> <span style="color: #0000FF;">=</span> <span style="color: #7060A8;">compare</span><span style="color: #0000FF;">(</span><span style="color: #000000;">s1</span><span style="color: #0000FF;">[</span><span style="color: #000000;">i1</span><span style="color: #0000FF;">],</span><span style="color: #000000;">s2</span><span style="color: #0000FF;">[</span><span style="color: #000000;">i2</span><span style="color: #0000FF;">])</span>
<span style="color: #000000;">res</span> <span style="color: #0000FF;">+=</span> <span style="color: #0000FF;">(</span><span style="color: #000000;">c</span><span style="color: #0000FF;">=</span><span style="color: #000000;">0</span><span style="color: #0000FF;">)</span>
<span style="color: #000000;">i1</span> <span style="color: #0000FF;">+=</span> <span style="color: #0000FF;">(</span><span style="color: #000000;">c</span><span style="color: #0000FF;"><=</span><span style="color: #000000;">0</span><span style="color: #0000FF;">)</span>
<span style="color: #000000;">i2</span> <span style="color: #0000FF;">+=</span> <span style="color: #0000FF;">(</span><span style="color: #000000;">c</span><span style="color: #0000FF;">>=</span><span style="color: #000000;">0</span><span style="color: #0000FF;">)</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">while</span>
<span style="color: #008080;">return</span> <span style="color: #000000;">res</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">function</span>
Line 72 ⟶ 643:
<span style="color: #004080;">sequence</span> <span style="color: #000000;">scores</span> <span style="color: #0000FF;">=</span> <span style="color: #0000FF;">{},</span>
<span style="color: #000000;">s1</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">bigram</span><span style="color: #0000FF;">(</span><span style="color: #000000;">s</span><span style="color: #0000FF;">)</span>
<span style="color: #004080;">integer</span> <span style="color: #000000;">l1</span> <span style="color: #0000FF;">=</span> <span style="color: #7060A8;">length</span><span style="color: #0000FF;">(</span><span style="color: #000000;">s1</span><span style="color: #0000FF;">)</span>
<span style="color: #008080;">for</span> <span style="color: #000000;">phrase</span> <span style="color: #008080;">in</span> <span style="color: #000000;">dictionary</span> <span style="color: #008080;">do</span>
<span style="color: #004080;">sequence</span> <span style="color: #000000;">s2</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">bigram</span><span style="color: #0000FF;">(</span><span style="color: #000000;">phrase</span><span style="color: #0000FF;">)</span>
<span style="color: #004080000000;">integerscores</span> <span style="color: #0000FF;">&=</span> <span style="color: #000000;">l22</span><span style="color: #0000FF;">*</span><span style="color: #000000;">intrasect</span><span style="color: #0000FF;">(</span><span style="color: #000000;">s1</span><span style="color: #0000FF;">,</span><span style="color: #000000;">s2</span><span style="color: #0000FF;">)/(</span><span style="color: #7060A8;">length</span><span style="color: #0000FF;">(</span><span style="color: #000000;">s1</span><span style="color: #0000FF;">)+</span><span style="color: #7060A8;">length</span><span style="color: #0000FF;">(</span><span style="color: #000000;">s2</span><span style="color: #0000FF;">))</span>
<span style="color: #000000;">s2</span> <span style="color: #0000FF;">=</span> <span style="color: #000000;">intersection</span><span style="color: #0000FF;">(</span><span style="color: #000000;">s2</span><span style="color: #0000FF;">,</span><span style="color: #000000;">s1</span><span style="color: #0000FF;">)</span>
<span style="color: #000000;">scores</span> <span style="color: #0000FF;">&=</span> <span style="color: #000000;">2</span><span style="color: #0000FF;">*</span><span style="color: #7060A8;">length</span><span style="color: #0000FF;">(</span><span style="color: #000000;">s2</span><span style="color: #0000FF;">)/(</span><span style="color: #000000;">l1</span><span style="color: #0000FF;">+</span><span style="color: #000000;">l2</span><span style="color: #0000FF;">)</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">for</span>
<span style="color: #7060A8;">printf</span><span style="color: #0000FF;">(</span><span style="color: #000000;">1</span><span style="color: #0000FF;">,</span><span style="color: #008000;">"%s &gt;\n"</span><span style="color: #0000FF;">,</span><span style="color: #000000;">s</span><span style="color: #0000FF;">)</span>
Line 84 ⟶ 652:
<span style="color: #7060A8;">printf</span><span style="color: #0000FF;">(</span><span style="color: #000000;">1</span><span style="color: #0000FF;">,</span><span style="color: #008000;">"%f %s\n"</span><span style="color: #0000FF;">,{</span><span style="color: #000000;">scores</span><span style="color: #0000FF;">[</span><span style="color: #000000;">t</span><span style="color: #0000FF;">],</span><span style="color: #000000;">dictionary</span><span style="color: #0000FF;">[</span><span style="color: #000000;">t</span><span style="color: #0000FF;">]})</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">for</span>
<span style="color: #7060A8;">printf</span><span style="color: #0000FF;">(</span><span style="color: #000000;">1</span><span style="color: #0000FF;">,</span><span style="color: #008000;">"\n"</span><span style="color: #0000FF;">)</span>
<span style="color: #008080;">end</span> <span style="color: #008080;">procedure</span>
Line 95 ⟶ 664:
Almkvist-Giullera formula for pi
Almost prime
Bell numbers
Check Machin-like formulas
Chowla numbers
Line 118 ⟶ 686:
<!--</syntaxhighlight>-->
{{out}}
Extending the task list to the full 1577 entries changes nothing.
You can get the same results as raku by setting match_raku to true, ie "uniqueify" all the bi-grams, so for instance "Primordial primes" adds "pr" and "pr1", whereas when match_raku is true it just adds "pr", once, and of course the same for "ri", "im", etc. Also, extending the task list to the full 1577 entries changes nothing either way.
<pre>
Primordial primes >
0.695652685714 FactorialSequence of primorial primes
0.642857666667 Sequence of primorialFactorial primes
0.571429 Primorial numbers
0.631579 Prime words
0.545455 Prime words
0.600000 Almost prime
0.521739 Almost prime
0.583333 Primorial numbers
 
Sunkist-Giuliani formula >
0.571429565217 Almkvist-Giullera formula for pi
0.352941378378 HaversineFaulhaber's formula
0.342857 Haversine formula
0.350000 Check Machin-like formulas
0.333333 Check Machin-like formulas
0.342857 Faulhaber's formula
0.315789307692 Resistance calculator
 
Sieve of Euripides >
0.461538 Sieve of Pritchard
Line 138 ⟶ 708:
0.400000 Piprimes
0.384615 Sierpinski curve
 
Chowder numbers >
0.818182782609 Chowla numbers
0.636364640000 RhondaPowerful numbers
0.631579608696 LahRhonda numbers
0.608696 PowerfulFermat numbers
0.571429600000 FermatLah numbers
</pre>
 
=={{header|Python}}==
Of the several Python string similarity libraries implementing Sorenson-Dice similarity, none give the same results as the
original example's Raku library, so this was imitated using Multisets, as per the C++ and Wren examples.
<syntaxhighlight lang="python">''' Rosetta Code task rosettacode.org/wiki/Sorensen–Dice_coefficient '''
 
from multiset import Multiset
 
 
def tokenizetext(txt):
''' convert a phrase into a count of bigram tokens of its words '''
arr = []
for wrd in txt.lower().split(' '):
arr += ([wrd] if len(wrd) == 1 else [wrd[i:i+2]
for i in range(len(wrd)-1)])
return Multiset(arr)
 
 
def sorenson_dice(text1, text2):
''' Sorenson-Dice similarity of Multisets '''
bc1, bc2 = tokenizetext(text1), tokenizetext(text2)
return 2 * len(bc1 & bc2) / (len(bc1) + len(bc2))
 
 
with open('tasklist_sorenson.txt', 'r') as fd:
alltasks = fd.read().split('\n')
 
for testtext in ['Primordial primes', 'Sunkist-Giuliani formula',
'Sieve of Euripides', 'Chowder numbers']:
taskvalues = sorted([(sorenson_dice(testtext, t), t)
for t in alltasks], reverse=True)
print(f'\n{testtext}:')
for (val, task) in taskvalues[:5]:
print(f' {val:.6f} {task}')
</syntaxhighlight>{{out}}
<pre>
Primordial primes:
0.685714 Sequence of primorial primes
0.666667 Factorial primes
0.571429 Primorial numbers
0.545455 Prime words
0.521739 Almost prime
 
Sunkist-Giuliani formula:
0.565217 Almkvist-Giullera formula for pi
0.378378 Faulhaber's formula
0.342857 Haversine formula
0.333333 Check Machin-like formulas
0.307692 Resistance calculator
 
Sieve of Euripides:
0.461538 Sieve of Pritchard
0.461538 Four sides of square
0.413793 Sieve of Eratosthenes
0.400000 Piprimes
0.384615 Sierpinski curve
 
Chowder numbers:
0.782609 Chowla numbers
0.640000 Powerful numbers
0.608696 Rhonda numbers
0.608696 Fermat numbers
0.600000 Lah numbers
</pre>
 
Line 189 ⟶ 824:
0.608696 Rhonda numbers
0.6 Lah numbers</pre>
 
=={{header|Wren}}==
{{libheader|Wren-str}}
{{libheader|Wren-set}}
{{libheader|Wren-fmt}}
This assumes that a one letter word is treated as a bigram. It also assumes that all bigrams are matched whether duplicates or not.
 
The results on this basis are the same as the Raku example.
<syntaxhighlight lang="wren">import "io" for File
import "./str" for Str
import "./set" for Bag
import "./fmt" for Fmt
 
var bigrams = Fn.new { |phrase|
var words = Str.splitNoEmpty(phrase, " ")
var res = []
for (word in words) {
var chars = Str.lower(word).toList
if (chars.count == 1) {
res.add(chars[0])
} else {
for (i in 0...chars.count-1) {
res.add(chars[i] + chars[i+1])
}
}
}
return res
}
 
var sorensen = Fn.new { |a, b|
var abi = Bag.new(bigrams.call(a))
var bbi = Bag.new(bigrams.call(b))
var common = abi.intersect(bbi)
return 2 * common.count / (abi.count + bbi.count)
}
 
var fileName = "rc_tasks_2022_09_24.txt" // local copy
var tasks = File.read(fileName).trimEnd().split("\n")
var tc = tasks.count
var tests = [
"Primordial primes", "Sunkist-Giuliani formula", "Sieve of Euripides", "Chowder numbers"
]
var sdis = List.filled(tc, null)
for (test in tests) {
for (i in 0...tasks.count) sdis[i] = [tasks[i], sorensen.call(tasks[i], test)]
var top5 = sdis.sort { |e1, e2| e1[1] >= e2[1] }.take(5).toList
System.print("%(test) >")
for (e in top5) Fmt.print(" $f $s", e[1], e[0])
System.print()
}</syntaxhighlight>
 
{{out}}
<pre>
Primordial primes >
0.685714 Sequence of primorial primes
0.666667 Factorial primes
0.571429 Primorial numbers
0.545455 Prime words
0.521739 Almost prime
 
Sunkist-Giuliani formula >
0.565217 Almkvist-Giullera formula for pi
0.378378 Faulhaber's formula
0.342857 Haversine formula
0.333333 Check Machin-like formulas
0.307692 Resistance calculator
 
Sieve of Euripides >
0.461538 Four sides of square
0.461538 Sieve of Pritchard
0.413793 Sieve of Eratosthenes
0.400000 Piprimes
0.384615 Sierpinski curve
 
Chowder numbers >
0.782609 Chowla numbers
0.640000 Powerful numbers
0.608696 Fermat numbers
0.608696 Rhonda numbers
0.600000 Lah numbers
</pre>
Peak
2,442

edits

Retrieved from "https://rosettacode.org/wiki/Special:MobileDiff/330605...361618"