Fuzzy Matching/string similarity in Qliksense

[Tan1]

Is fuzzy matching possible in Qliksense with 60-70% string similarity? 

For Eg. 

I have 2 tables. 

Feild1:

F1

From Source1; 

Field2:

F2

from Source2; 

Suppose in Field1, I am having 'IBM & SALARY COMPLETE PROJECTS'  and in Field2 I will be having 'IBM INDIA PVT LTD'. I want to join/link Field1 and Field2 with 60% similarity between the strings. 

Please let me know if it's feasible in QlikSense. 

[JandreKillianRIC]

Hi @Tan1 

You can have a look at LevenshteinDist. You might just need to build a key using some of these values along with something like Autonumber. 

LevenshteinDist() returns the Levenshtein distance between two strings. It is defined as the minimum number of single-character edits (insertions, deletions, or substitutions) required to change one string into the other. The function is useful for fuzzy string comparisons.

Regards Jandre

[Tan1]

But I have two separate tables. How can I create association using LevenshteinDist() function? 

[JandreKillianRIC]

Hi @Tan1 

I would join the 2 tables, On a resident load have both the fields, F1, F2 and the LevenshteinDist value. 

Then using the above table to look at the values you want to join based on a match and join that, 

If you can provide a bit more data I can build the back end and send it to you. 

Regards - Jandre

[Tan1]

Hi, 

I have attached dummy data for your reference. 

[JandreKillianRIC]

Hi @Tan1 

Here is a qvf that links the 2. Let me know if you need this in any other way. (The distance table acts as a small link table - For a lack of a better word) 

Regards - Jandre

[Dana_Baldwin]

Hi @Tan1 

Please note the forum you have posted to is for the Data Integration product Enterprise Manager, used for managing multiple Replicate/Compose/Catalog servers. To reach your target audience, please consider posting your question here: https://community.qlik.com/t5/Qlik-Sense/ct-p/qlik-sense

Thanks,

Dana