jaro-winkler

I've been wondering for months about how to implement this algorithm in Transact SQL, https://en.wikipedia.org/wiki/Jaro%E2%80%93Winkler_distance How can it be done? Maritim Today I finally stumbled upon this Stack Overflow-answer by leebickmtu showing an implementation in C#, originally ported from Java. I took the liberty to port it to a Transact SQL function, enjoy! IF OBJECT_ID (N'dbo.InlineMax', N'FN') IS NOT NULL DROP FUNCTION dbo.InlineMax; GO CREATE FUNCTION dbo.InlineMax(@valueOne int, @valueTwo int) RETURNS FLOAT AS BEGIN IF @valueOne > @valueTwo BEGIN RETURN @valueOne END RETURN

Is there any LGPL or commercial-friendly licensed implementation of Jaro-Winkler distance in .NET? The SimMetrics library appears to support Jaro-Winkler, and there's a .NET version available for download. Unfortunately it's licensed under the GPL, but maybe the authors would be amenable to giving/selling you a commercial license. (NB: I haven't used this library myself, and know absolutely nothing about it.) I wrote a public domain version in F#. You can access it here: http://fssnip.net/2M 来源： https://stackoverflow.com/questions/1510593/jaro-winkler-distance-algorithm-in-net

I have a use case where I need to do fuzzy matching of millions of records from multiple files. I identified two algorithms for that: Jaro-Winkler and Levenshtein edit distance. When I started exploring both, I was not able to understand what the exact difference is between the two. It seems Levenshtein gives the number of edits between two strings, and Jaro-Winkler gives a matching score between 0.0 to 1.0. I didn't understand the algorithm. As I need to use either algorithm, I need to know the exact differences with respect to algorithm performance. Levenshtein counts the number of edits