What are the techniques in handling categorical attributes?
How do continuous attributes differ from categorical attributes?
What is a concept hierarchy?
Note the major patterns of data and how they work.
One-hot encoding is the commonest way of handling non-ordinal categorical data. It involves creating an extra feature for every group of the categorical characteristic and marking every observation belonging to the group. Target encoding is the other means of handling such data. Categorical attributes have a finite number of categories, while continuous variables contain an infinite number of values between any two figures. A concept hierarchy describes a series of mappings from low-level to higher-level concepts that are usually more general.