in src/main/java/org/apache/commons/text/similarity/JaccardSimilarity.java [69:90]
private Double calculateJaccardSimilarity(final CharSequence left, final CharSequence right) {
final int leftLength = left.length();
final int rightLength = right.length();
if (leftLength == 0 && rightLength == 0) {
return 1d;
}
if (leftLength == 0 || rightLength == 0) {
return 0d;
}
final Set<Character> leftSet = new HashSet<>();
for (int i = 0; i < leftLength; i++) {
leftSet.add(left.charAt(i));
}
final Set<Character> rightSet = new HashSet<>();
for (int i = 0; i < rightLength; i++) {
rightSet.add(right.charAt(i));
}
final Set<Character> unionSet = new HashSet<>(leftSet);
unionSet.addAll(rightSet);
final int intersectionSize = leftSet.size() + rightSet.size() - unionSet.size();
return 1.0d * intersectionSize / unionSet.size();
}