Identifying to which category an object belongs to.

Predicting a continuous-valued attribute associated with an object.

Automatic grouping of similar objects into sets.

Reducing the number of random variables to consider.

Comparing, validating and choosing parameters and models.

Feature extraction and normalization.