Sorensen–Dice coefficient: Difference between revisions

Content added Content deleted

Inline

@@ Line 57: / Line 57: @@
 #include <vector>
+using bigram = std::pair<char, char>;
-std::multiset<std::string> split(const std::string& phrase) {
-    std::multiset<std::string> result;
+std::multiset<bigram> split(const std::string& phrase) {
+    std::multiset<bigram> result;
     std::istringstream is(phrase);
     std::string word;
@@ Line 67: / Line 69: @@
         size_t length = word.size();
         if (length == 1) {
-            result.emplace(1, word[0]);
+            result.emplace(word[0], '\0');
         } else {
             for (size_t i = 0; i + 1 < length; ++i) {
-                result.insert(std::string{word[i], word[i + 1]});
+                result.emplace(word[i], word[i + 1]);
             }
         }
@@ Line 80: / Line 82: @@
     auto a = split(s1);
     auto b = split(s2);
-    std::multiset<std::string> c;
+    std::multiset<bigram> c;
     std::set_intersection(a.begin(), a.end(), b.begin(), b.end(),
                           std::inserter(c, c.begin()));