Standardization (Z-score)

Description

Z-score normalization refers to the process of normalizing every value in a dataset such that the mean of All the values is 0 and the standard deviation is 1.

This technique transforms the feature values to have a mean of 0 and a standard deviation of 1. Standardization is less affected by outliers in the data than min-max scaling.

Formula

• \(Z\) = standard score
• \(x\) = observed value
• \(\mu\) = mean of the sample
• \(\sigma\) = standard deviation of the sample

Example

مقدار \(\sigma\) برابر است با انحراف از معیار (standard deviation)