What is the name of the process that converts a continuous variable into a discrete variable?

The process that converts a continuous variable into a discrete variable is known as discretization. Discretization involves taking a range of continuous data points and grouping them into distinct categories or bins. This is often done to simplify models, reduce the complexity of the data, or make the data more interpretable. In many machine learning applications, discretizing continuous variables can help in algorithms that perform better with categorical data rather than numerical input.

While the terms transformation, normalization, and standardization refer to specific data preprocessing techniques, they do not specifically describe the conversion of continuous variables into discrete ones. Normalization typically refers to scaling the data to a specific range, especially when preparing data for algorithms that assume a specific range of input values. Standardization involves centering the data around the mean with a standard deviation of one, which helps in cases where the data follows a Gaussian distribution. Transformation is a broader term that includes any function applied to data, but it does not imply discretization directly.