I was wondering if it makes sense to apply normal standardization on a feature like timestamp ? The data that I process are network packets.

Thank you

  • $\begingroup$ Could you expand on what you want to use this feature to do? $\endgroup$
    – jamesmf
    Nov 5, 2015 at 16:18

1 Answer 1


For time series analysis. Yes

But, turning data into a computable object for using in the ML computation? No

Using the data as a feature. Then, Yes

I would give you a general time series example:

Consider the number of days in months. The irregularity causes friction while analyzing the model.

Consider this: enter image description here

So, such type of transformations would be helpful which analyzing time series', which can help reduce friction in the models sensibly. Link of the explanation


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.