August 4, 2020

NORMALIZATION

1. The goal of normalization is to make every datapoint have the same scale so each feature is equally important.

2. Min-Max Normalization is one of the most common ways to normalize data. For every feature, the minimum value of that feature gets transformed into a 0, the maximum value gets transformed into a 1, and every other value gets transformed into a decimal between 0 and 1.

3. Z-score Normalization is a strategy of normalizing data that avoids the issue of outliers.

   1. (value - u / o

   2. Here, u is the mean value of the feature and o is the standard deviation of the feature.

      1. The mean will be 0 and anything below the mean will be negative, while anything above the mean will be positive.

4. Min-Max guarantees the same scale, but poorly handles outliers.

5. Z-score handles outliers, but not on the same scale.