Levenshtein Distance Calculator

Work out levenshtein distance instantly with clear inputs, formula shown and shareable results.

About Levenshtein Distance Calculator

Levenshtein distance is the minimum number of single-character insertions, deletions or substitutions needed to turn one string into another, computed with a dynamic-programming table in O(mn) time. Dividing by the longer length gives a similarity ratio that is comparable across string lengths, which is what fuzzy matching, spell checking and record linkage use as a threshold.

Formula

Edit distance recurrence

D(x,y) = min(D(x-1,y) + 1, D(x,y-1) + 1, D(x-1,y-1) + cost) where cost is 0 when the characters match

Worked example

With the default values below, Levenshtein Distance Calculator returns Levenshtein distance of 3. Change any input above and every figure here is recalculated in your browser.

Inputs used

First string: kitten
Second string: sitting

Results

Levenshtein distance: 3
Normalised similarity: 57.14%
Length of the longer string: 7
Normalised distance: 0.4286

Step by step

1Values used
2Edit distance recurrence
3Levenshtein distance= 3
4Normalised similarity= 57.14
5Length of the longer string= 7
6Normalised distance= 0.4286

Assumptions and limits

The result is derived only from the 2 inputs listed above. Anything not among them is not part of the calculation.
In particular, this calculator does not take inflation, tax, fees or charges or compounding frequency into account.
Figures are rounded for display to 4 decimal places, so totals built from several rounded parts can differ slightly from the unrounded arithmetic.

How these calculators are built and tested is described on the Methodology page. If a figure looks wrong, please report it.

Frequently Asked Questions

How does this differ from Damerau-Levenshtein?

Damerau-Levenshtein also allows transposing two adjacent characters as a single edit, which better reflects typing errors such as teh for the.

What similarity threshold works for record matching?

Around 0.85 to 0.9 normalised similarity is a common starting point for names, but always calibrate against a labelled sample because it depends heavily on field length.