Normalization and Standardization

Normalization and Standardization
0

Hi,

In Week1 : Introduction “Processing Data” it was mentioned that Normalization means to transform data into zero mean or unit variance and Standardization means to rescale and shift data values so that they become between 0 and 1.

However, on searching net for the same it shows vice-versa.
Please find below one link:

Can anyone please clarify?

Regards,
Neha

Normalization rescales the values into a range of [0,1]. This might be useful in some cases where all parameters need to have the same positive scale. However, the outliers from the data set are lost.

Standardization rescales data to have a mean (μ) of 0 and standard deviation (σ) of 1 (unit variance).

“Standardization” typically means that the range of values are “standardized” to measure how many standard deviations the value is from its mean.

For most applications standardization is recommended.

This might be helpful.

1 Like

Yes there’s a typo over there in the slides… will put a disclaimer over there…

Yes that’s correct. But what is mentioned in the slide is the other way round.