Automatic Classification and Intelligent Clustering for WWWeb Information Retrieval Systems

In this paper we present some aspects of an intelligent interface for a WWWeb legal information retrieval system. Our system is able to keep the context of the user interaction in order to supply suggestions for further refinement of the user queries. The set of documents obtained from the user queries is dynamically organised in clusters labeled with keywords from a juridical thesaurus. Since, some of the texts were not previously classified, we have developed an automatic juridical classifier based on a neural network. The classifier receives as input a legal text and proposes a set of juridical terms that characterize it.