Best Answer · By JanBask Data Science Expert
Answered on Nov 30, 2019
Normalization refers to rescaling real valued numeric attributes into the range 0 and 1.
It is useful to scale the input attributes for a model that depends on the magnitude of values, such as distance measures algorithm used in models like k-nearest neighbors or clustering and in the preparation of coefficients in regression.
It works on the following formula
The below graph shows the difference between original and normalized data.
Was this helpful?
