July 2017
Beginner to intermediate
486 pages
13h 49m
English
The statistical aspect of normalization is used to do features scaling. If you have a dataset where one data attribute's ranges are too high and the other data attributes' ranges are too small, then generally we need to apply statistical techniques to bring all the data attributes or features into one common numerical range. There are many ways to perform this transformation, but here we will illustrate the most common and easy method of doing this called min-max scaling. Let's look at equation and mathematical examples to understand the concept.
Min-max scaling brings the features in the range of [0,1]. The general formula is given in Figure 5.65:
Read now
Unlock full access