Skip to main content Link Search Menu Expand Document (external link)
Feature scaling via data normalization is a popular method of input normalization in Machine Learning. It provides several benefits to the learning process. In fact, it improves the learning speed when a local optimization algorithm is used. This technique mainly consists of first mean-centering and the resclaing each of the input features by the inverse of its standard deviation. This simple technique can significantly improve the data analysis and provide better modeling insights.

normalizationCostEffect

Feature Engineering: Effect of data normalization