A Methodology based on Commonsense Knowledge and Ontologies for the Automatic Classification of Legal Cases

We describe a methodology for the automatic classificationof legal cases expressed in natural language, which relieson existing legal ontologies and a commonsense knowledgebase. This methodology is founded on a process consisting ofthree phases: an enrichment of a given legal ontology by as-sociating its terms with topics retrieved from the Wikipediaknowledge base; an extraction of relevant concepts from agiven textual legal case; and a matching between the en-riched ontological terms and the extracted concepts. Such aprocess has been successfully implemented in a correspond-ing tool that is part of a larger framework for self-litigationand legal support for the Italian law.