Filtering Values in a Group_ID

[Sarath27]

Hi All,

Please help me guiding to create a workflow for the sample data attached.

Description

Sample data attached were grouped by FULL_NAME,given Group_ID for the same.

Requirement

1. Each Group_ID has set of records, which any one of them has "5 CANADA SQUARE" in ADDR_LINE1 column, *but the requirement is* If the other addresses in the same group is Same/duplicate, then it should be labelled as "2".

2. If other addresses in the same group is Unique, then it should be labelled as "3".

Example for your reference ( And also I added detailed sample data use that as well if you need)

Group_ID	FULL_NAME	ADDR_LINE1	Label (Required Output)
1	S	5 CANADA SQUARE	2
1	S	Aiy	2
1	S	Aiy	2
2	B	Biy	3
2	B	5 CANADA SQUARE	3
2	B	Dye	3
3	S	Cyr	2
3	S	5 CANADA SQUARE	2
3	S	Cyr	2
4	A	Temple	3
4	A	Temple	3
4	A	5 CANADA SQUARE	3
4	A	Buel	3

[IraWatt · Accepted solution]

Hey @Sarath27,

I've attached a workflow which I think captures this logic:

If have any questions on how it works or any issues make sure to ask :) 

HTH,

Ira

[IraWatt]

Two questions I have is

why is cluster 76 labelled 3 in your dataset?

And cluster 82 has a mixture of labels?

[Sarath27]

Hi @IraWatt  

Thanks much for your prompt response. Both the Clusters 72 &86 should be labelled as "2".  I labelled it manually so it might have been typed erroneously.

[IraWatt]

Hey @Sarath27,

No worries, Glad it worked ! :)