(a) Supervised learning in which classes are
distinguished by word patterns:
An example can be that the documents in the
class China tend to have high values on dimensions
like Chinese, Beijing, Shanghai and Mao Tse Tung
whereas documents in the class India tend to have high
values for New Delhi, Gandhi and Mumbai. Enhancing
the effectiveness of classifiers has been the main
focus of research and a range of new techniques
(support vector machines, neural networks, etc.)
has been developed.