Fuzzy logic in SQL

Hi experts, I need to calculte the match percentage of these two, is there a way to do it?
You cannot do that with T-SQL, you will have to resort to writing a CLR function to accomplish that.

What do those names reference? I.e., what is the meaning of the data and what is the data definition?

Tom

Please provide with sample data and logic you want to use

Please look at the attached sample data, and I have the logic in place to figure out the exact matches but I  want the logic to see if there are any close matches in terms of percentage -
For ex: DescrCode 1 and 2 has exact match Description1 and Description2 , DescCode 3 - 6 match > 50% , 7 and 8 less than 50% , this is just tentative percentages but can be more definitive.
Please let me know if you need more info, thanks
sample-data.txt

this is the correct sample file, pls ignore previous one, thanks
sample-data.txt

thanks!

Well there are different ways of doing it e.g

1. As  acperkins said, you can use CLR function to accomplish and the example is
http://anastasiosyal.com/POST/2009/01/11/18.ASPX?

2. As liija said, use SSIS package to do the job

3. If you have SQL Server 2008 R2 then a another option is to install Master Data Services
http://msdn.microsoft.com/en-us/library/ee633742(v=sql.105).aspx

sample code will be something like

DECLARE 
	@Name VARCHAR(100), 
	@Name2 VARCHAR(100)

SET @Name = 'rehan yousaf'
SET @Name2 = 'rayhaan usuf'

SELECT 
	@Name AS [@Name], 
	@Name2 AS [@Name2],
    MDS.mdq.Similarity(@Name, @Name2, 0, 0.85, 0) AS Levenshtein,
    MDS.mdq.Similarity(@Name, @Name2, 1, 0.85, 0) AS Jaccard,
    MDS.mdq.Similarity(@Name, @Name2, 2, 0.85, 0) AS JaroWinkler,
    MDS.mdq.Similarity(@Name, @Name2, 3, 0.85, 0) AS LongestCommonSubsequence

Select all Open in new window

I have also attached the screenshot for option 3 as well
FuzzyLogic01.JPG