Jaro–Winkler distance

This article is about the measure. For other uses, see Jaro.

In computer science and statistics, the Jaro–Winkler distance (Winkler, 1990) is a measure of similarity between two strings. It is a variant of the Jaro distance metric (Jaro, 1989, 1995), a type of string edit distance, and was developed in the area of record linkage (duplicate detection) (Winkler, 1990). The higher the Jaro–Winkler distance for two strings is, the more similar the strings are. The Jaro-Winkler similarity (for equation see below) is given by 1 - Jaro Winkler distance. The Jaro–Winkler distance metric is designed and best suited for short strings such as person names. The score is normalized such that 0 equates to no similarity and 1 is an exact match.

Definition

The Jaro distance d_j of two given strings s_1 and s_2 is

d_j = \left\{

\begin{array}{l l}
  0 & \text{if }m = 0\\
  \frac{1}{3}\left(\frac{m}{|s_1|} + \frac{m}{|s_2|} + \frac{m-t}{m}\right) & \text{otherwise} \end{array} \right.

Where:

Two characters from s_1 and s_2 respectively, are considered matching only if they are the same and not farther than \left\lfloor\frac{\max(|s_1|,|s_2|)}{2}\right\rfloor-1.

Each character of s_1 is compared with all its matching characters in s_2. The number of matching (but different sequence order) characters divided by 2 defines the number of transpositions. For example, in comparing CRATE with TRACE, only 'R' 'A' 'E' are the matching characters, i.e. m=3. Although 'C', 'T' appear in both strings, they are farther than 1, i.e., floor(5/2)-1=1. Therefore, t=0 . In DwAyNE versus DuANE the matching letters are already in the same order D-A-N-E, so no transpositions are needed.

Jaro–Winkler distance uses a prefix scale p which gives more favourable ratings to strings that match from the beginning for a set prefix length \ell. Given two strings s_1 and s_2, their Jaro–Winkler distance d_w is:

d_w = d_j + (\ell p (1 - d_j))

where:

Although often referred to as a distance metric, the Jaro–Winkler distance is actually not a metric in the mathematical sense of that term because it does not obey the triangle inequality . In fact the Jaro–Winkler distance also does not satisfy that axiom that states that   d(x,y)=0 \rightarrow x = y.

In some implementations of Jaro-Winkler, the prefix bonus \ell p (1 - d_j) is only added when the compared strings have a Jaro distance above a set "boost threshold" b_t. The boost threshold in Winkler's implementation was 0.7.

d_w = \left\{

\begin{array}{l l}
  d_j & \text{if }d_j < b_t\\
  d_j + (\ell p (1 - d_j)) & \text{otherwise} \end{array} \right.

Example

Note that Winkler's "reference" C code differs in at least two ways from published accounts of the Jaro–Winkler metric. First is his use of a typo table (adjwt) and also some optional additional tolerance for long strings.

Example #1

Given the strings s_1 MARTHA and s_2 MARHTA we find:

We find a Jaro score of:

d_j = \frac{1}{3}\left(\frac{6}{6} + \frac{6}{6} + \frac{6-1}{6}\right) = 0.944

To find the Jaro–Winkler score using the standard weight p = 0.1, we continue to find:

\ell = 3

Thus:

d_w = 0.944 + (3 * 0.1 (1 - 0.944)) = 0.961

Given the strings s_1 DWAYNE and s_2 DUANE we find:

We find a Jaro score of:

d_j = \frac{1}{3}\left(\frac{4}{6} + \frac{4}{5} + \frac{4-0}{4}\right) = 0.822

To find the Jaro–Winkler score using the standard weight p = 0.1, we continue to find:

\ell = 1

Thus:

d_w = 0.822 + (1 * 0.1 (1 - 0.822)) = 0.84

Example #2

Given the strings s_1 DIXON and s_2 DICKSONX we find:

D I X O N
D 1 0 0 0 0
I 0 1 0 0 0
C 0 0 0 0 0
K 0 0 0 0 0
S 0 0 0 0 0
O 0 0 0 1 0
N 0 0 0 0 1
X 0 0 0 0 0

Here, the shaded cells are the match window for each character. A 1 in a cell indicates a match. Note that the two Xs are not considered matches because they are outside the match window of 3.

We find a Jaro score of:

d_j = \frac{1}{3}\left(\frac{4}{5} + \frac{4}{8} + \frac{4-0}{4}\right) = 0.767

To find the Jaro–Winkler score using the standard weight p = 0.1, we continue to find:

\ell = 2

Thus:

d_w = 0.767 + (2 * 0.1 (1 - 0.767)) = 0.814

See also

References

External links

This article is issued from Wikipedia - version of the Thursday, April 07, 2016. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.