Hello everybody,
i'm sorry for my English(i'm from Czech Republic), but i have important question for my job.
I did in Google Refine keyword analysis and i used clustering to rename similar rows. But i have in next cells data about search count from adwords and i need sum up search count in duplicate rows now(duplicate in keyword, but different in search count).
Is this possible in Google Refine?
Hi Filip. I'm sure your English is much better than my Czech!
If your data looks like:
Keyword Count
keyword1 10
keyword1 15
keyword2 42
keyword2 51
what you want to do is:
- Blank Down command on the Keyword column
- move the Keyword colum to the first (left) position if it's not already there
- Join Multi-valued Cells on the Count column
- Add a column called Total Count with the expression: forEach(value.split(','),v,v.toNumber()).sum()
This takes advantage of OpenRefine's concept of "records" which it forms based on the missing initial column values after the Blank Down operation. You can switch between Record and Row display mode to get different effects depending on what you are trying to do.
Tom