In this video, we will learn what data scaling is, why it is important and its basic algorithm with example. Data scaling is the most important pre-processing step when working with deep learning neural networks. It can be achieved by normalizing or standardizing with real-valued input and output variables. Normalizing is min-max scaling which computes the minimum and maximum value of features are used for scaling. And standardizing is used to Mean and the standard deviation for scaling. In general, Data scaling is a preprocessing technique that needs to be performed on the data before passing it into any machine learning model. 

Reference Articles: 

https://www.geeksforgeeks.org/ml-feature-scaling-part-1/ 

https://www.geeksforgeeks.org/ml-feature-scaling-part-2/ 

https://www.geeksforgeeks.org/feature-scaling-part-3/