In data transformation, what is the primary purpose of standardization?

Standardization is a data transformation technique that resizes the distribution of values to have a mean of 0 and a standard deviation of 1. This approach is beneficial when data has different units or ranges, as it aligns them for fair comparison and analysis. Standardization is commonly used in machine learning algorithms that rely on Gaussian distributions, helping improve model performance by normalizing feature scales. Option A is incorrect as scaling data to a range between 0 and 1 is normalization, not standardization. Option B is incorrect because converting categorical to numerical data is encoding, a separate transformation technique. Option D is incorrect as removing outliers is part of data cleaning, not standardization. Option E is incorrect because standardization does not involve data partitioning for training and testing.