**Ordinal Encoding**: This is another way to convert a categorical variable into a numeric representation suitable for model training. Unlike dummy encoding, ordinal encoding requires knowledge of the target values corresponding to each observation. It works by sorting the original data in ascending order by the values of the target and then replacing the raw values with the index of the sorted data. An advantage of ordinal encoding is that it creates a monotonic relationship between the transformed input variable and target, which is beneficial for some algorithms. It is important to obtain the mapping from the original to transformed version of the variable using only the training data to avoid information from the test set leaking into the training process and possibly inducing bias.