Categorical Data preprocessing for Data mining

1 view (last 30 days)
Samuel Katongole
Samuel Katongole on 6 Oct 2021
Edited: Samuel Katongole on 6 Oct 2021
Hello friends
I have been working on the Tanzania wells state ,with Taarifa data obtained from DrivenData, problem for my ML practice; and I am now trying to remove misspellings in the installer and funder columns. Anyone who's tried this to please help me on how to go about it. And if there be a faster way, that would be very helpful.
Oh, thanks
I am trying to clean out misspellings from the installer and funder columns. For the moment I am using regular expressions; though the data is too much, and seems to be taking longer.
For instance, when trying to correct those for world bank I tried this expression which is still failing
pat11='wo(rd|rdl|uld|rld)?\s((b\w*|nk|divisio)$)?[^vd]';
newDataClean.installer=regexprep(newDataClean.installer,pat11,'world bank');
Here i was testing the expression in Atom, but it fails to correctly replace those selected words
However, I am still wondering if there could be another "faster" way of approaching the issue!
  1 Comment
KSSV
KSSV on 6 Oct 2021
Question is not clear. Can you elaborate with an example?

Sign in to comment.

Answers (0)

Categories

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!