Text levenshtein

Levenshtein

The edit distance most people mean when they say "how close are these two strings".

Unit costs, so every insertion, deletion and substitution counts one. Reach for Distance when you want a count you can threshold, and for NormalizedSimilarity when you want a score comparable across pairs of different lengths. A swap of two neighbouring characters costs two edits here; when that is the mistake you are chasing, DamerauLevenshtein is the one that charges one.

Members

Member	What it does
Levenshtein.Distance	Counts the fewest insertions, deletions and substitutions that turn one string into the other.
Levenshtein.NormalizedDistance	Scales the distance into [0, 1] by dividing it by the length of the longer input.
Levenshtein.NormalizedSimilarity	1 - NormalizedDistance: 1 when the two are identical, 0 when nothing survives.