dear community ,
my dataframe contain 24 column including lead id and user id , in fact I'm trying to analyze the dataframe in order to understand thee relationship between columns and its importance in parallel with the output , while analyzing I thought about deleting the 2 columns (user id and lead id ) but then I realized that it can't be possible cause there is some informations inside like a user can have more than 1 lead so the id user can be repeated more than once in different raws , my question is how to handle this ? how to analyse id data ? thank you
Dear Wassim Diai
These stack flow discussion might be useful, have a look:
https://stackoverflow.com/questions/41125138/create-column-named-id-based-on-row-index
Kind Regards
Qamar Ul Islam
You may try this rcode to generate a new unique ID combining two other IDs:
mydata %>%
dplyr::mutate(unique_ID= group_indices(lead_id, lead_id))
For more please visit this link:
"R - Group by variable and then assign a unique ID - Stack Overflow" https://stackoverflow.com/questions/39650511/r-group-by-variable-and-then-assign-a-unique-id
Hi Wassim Diai
It's not clear to me what the problem is, but I would try to provide a simple solution based on what I understand. Mutating the dataframe is an excellent suggestion made by Proloy Barua. An easy concept would be to merge lead ids and user ids into one column as IDs. The next step would be to remove duplicate rows if any exist. Rather than removing the ID column, you can simply use it as the row names. I could be more precise if you provide an example. Thanks
Dear Jagajjit Sahu , thank you for your answer !
let me please share a screenshot of my dataframe , as you can see we have the first 2 columns representing the lead id and user id , a user can have different leads , hence same id can be repeated , while processing the data and trying to calculate correlation or trying to make any feature Engineering or exploration on my data , I'm comfused about how to make the 2 columns useful for my model .
Thank you again !
Dear Wassim Diai
You can simply combine all your ID values into a single column and use it as row names, as I had already suggested. Using unique row-wise data allows you to conduct analysis such as correlation or whatever you desire. Take a look at the following links, for example. Thanks
https://stackoverflow.com/questions/20344779/combining-first-two-columns-and-turn-it-into-row-names-in-r-data-frame
https://www.dataanalytics.org.uk/combine-multiple-columns-as-row-names-in-r/
- Badges
- Science topic