The categorisation of hidden Web databases through concept specificity and coverage

Hidden Web databases maintain a collection of specialised documents, which are dynamically generated in response to users´ queries. The categorisation of such databases into a set of predefined categories has been widely employed to assist users in their information searches. In this paper we present a technique that automatically categorises a document database through its content summary and concepts described by their specificity and coverage. Experimental results show that our approach categorises databases with a larger number of relevant categories.