0
$\begingroup$

I've read a few vague articles and watched a couple of YouTube videos on data integrity and data sanity, but none of them have mentioned ways to actually check these on datasets.

I am interested in knowing the steps to check such factors on datasets.

Kindly let me know if this is too broad, I can think of more specific questions.

| improve this question | |
$\endgroup$
1
$\begingroup$

Here are a few things to check in a data set, there could be a lot more but i'm sharing what i did myself.

  • Check the NULLs in your columns, count them to see if there is a column that contains too many NULLs and that you might want to eliminate

  • Plot your data distribution each column at a time to see the range of values taken by your data, to see if there is something particularly remarquable about its behavior.

  • And I'd recommend this famous kernel on Kaggle where a good exploratory data analysis is done.
| improve this answer | |
$\endgroup$
1
$\begingroup$

You could check your dataset for:

  • Null values/Missing values
  • Zero values
  • Negative values
  • Max & Min values for each column
  • Checks for Date values if there are any in your dataset, if these values make sense.
| improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.