An iterative approach for lexicon characterization in juridical context

In the juridical context, knowledge management applications have a central role. In order to improve the effectiveness of document management procedures, techniques for automatic comprehension of textual content are required. In this work, a methodology for semi-automatic derivation of knowledge from document collections is proposed. In order to extract relevant information from document text, a process integrating both statistical and lexical approaches is applied. Moreover, we propose a system for the evaluation of the extracted peculiar lexicon quality. The system is used for the processing of heterogeneous documents corpus issued by Italy’s juridical domain.