Scribd
Upload
29 views2 pages

Standardization vs Normalization in ML

Standardization centers data around a mean of 0 and standard deviation of 1, maintaining the original data distribution and relative distances between points. Normalization scales data values to a fixed range like 0 to 1, preserving the original data shape but changing the distribution. Standardization is suitable for algorithms sensitive to variance or assuming normality, while normalization is for methods requiring values in a specific range.

Uploaded by

ymrodriguezg.5
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
29 views2 pages

Standardization vs Normalization in ML

Standardization centers data around a mean of 0 and standard deviation of 1, maintaining the original data distribution and relative distances between points. Normalization scales data values to a fixed range like 0 to 1, preserving the original data shape but changing the distribution. Standardization is suitable for algorithms sensitive to variance or assuming normality, while normalization is for methods requiring values in a specific range.

Uploaded by

ymrodriguezg.5
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Standardization

vs.
Normalization
Mohsen Kazemian
Standardization Normalization

Centers data around Scales data to a fixed


Goal zero (mean = 0) and range, often between
standard deviation = 1. 0 and 1.

Mean becomes 0, Range depends on the


Mean and Variance Standard deviation specified range (e.g.,
becomes 1 0 to 1)

Alters distribution but Keeps the original


maintains relative shape of the
Distribution
distances between distribution while
data points. scaling values.

Algorithms sensitive to Algorithms requiring


feature variance and values in a specific
Suitable for
assuming normal range or distance-
distribution. based computations.

Principal Component
Neural networks, k-
Analysis (PCA),
Example Use Case nearest neighbors,
algorithms like Linear
image processing.
Regression.

You might also like