What is normalization and how does it scale data within a specific range? How is standardization different in terms of centering data around the mean and adjusting variance? In which situations is normalization preferred over standardization, and vice versa? What are the common formulas used for both techniques? How do normalization and standardization impact the performance of machine learning models?