Fuzzy match but only keep 1 source

[MZ900605]

Hi Team am having a bit of trouble using Fuzzy matching, The first table is Master source contain clean data and way more records than i need.

The second table contain the names but some of them are written with mistakes, I need to match them and add the Lcodes to the second table but only the rows that is in user table. I tried ended up filling user table with all Master records. Any help guys?

ID MASTER	Names	L_codes
6y	Sam leo	324532453
7y	Monica freddy	657568342
8y	Jaye kkk	978686757
9y	Luke QQQ	211324345

ID USER	Names	Codes
16	Sam Loe	AUS22
17	Monica fredy	AUH00
18	Lukee QQQ	AOK33

[atcodedog05]

Hi @MZ900605 

Here is how you can do it.

Workflow:

Hope this helps : )

[kmedici]

Hi @MZ900605 

I've had a go at putting your use case into a workflow. Please have a look and let me know if it's what you were after! I've annotated the steps I took to achieve this results.

Thanks
Kris

[MZ900605]

@atcodedog05 this method is producing like 10 million records don't know why... and am using 92% fuzzy accuracy.

[atcodedog05]

Hi @MZ900605 

The issue might be many to many joins on similar names. Can you tell me number of input rows. Can you click on make group tool and check how many rows you are getting.

[MZ900605]

@kmedici when am applying union the source column "master" is replaced by NULL but names are shown.

names	codes	Source	L_codes
xzz	324123	NULL	32sdasd
xzzz	124123123	NULL	xyz

[MZ900605]

@atcodedog05 this what am having in this method.

[MZ900605]

@atcodedog05 the number of records are less than 10,000 the result is giving whole millions.but fuzzy matching gave the exact number all is good

[atcodedog05]

Hi @MZ900605 

Can you enable connection progress and show me the record flow.

[MZ900605]

@atcodedog05