Identify the type of data and the normalization goal
Remove or handle missing values
Detect and treat outliers if needed
Choose a normalization method
Min-max scale values to a fixed range such as 0 to 1
Standardize values to have mean 0 and standard deviation 1
Apply log transformation for highly skewed data
Apply decimal scaling when appropriate
Normalize each feature independently
Use the same fitted parameters on training and test data
Verify the normalized data distribution
Save the normalization parameters for future use