Research Project: JurWordNet, law semantic lexicon
More info

Semantic nets are means for content management. Coupled by Natural Language Processing instruments, they integrate parsing of morpho-syntactic aspects with meaning understanding, and are thus basic elements in many implementation areas, among which:

• support to information retrieval: although the recent research development in the semantic Web field has shifted focus on the research for information on basis structured for information extraction from non-structured data, WordNet was developed as a support to information retrieval systems, in order to ease the access to heterogeneous and multi-language data, through query extension and specialization.

• source of metadata for information extraction, automatic classification, and semantic tagging of legal documents. Many European countries have launched institutional projects for the definition of identification standards for legal sources, in particular for legislative texts. In England the project was developed by LAMS (Legal and Advice Sector Metadata): http://www.lcd.gov.uk/consult/meta/metafr.htm#part6). In Italy, by the project Norme in Rete (http://www.normeinrete.it/), coordinated by the Ministry of Justice. In the Netherlands by the project Metalex, at the University of Amsterdam (http://www.metalex.nl/pages/welcome.html). As for the international initiatives, the German Lexml project (http://www.lexml.de/), and the US one, Legalxml, (http://www.legalxml.org).

In Italy the Norme in Rete project has developed standard definitions (DTD XML) for the description of the typology and structure of texts in which the description of the contents is, for the time being, missing, or diffused through international resources (classification schemes, key words, etc). These may be integrated with semantic lexicons such as JurWordNet, tailored on the information needs of common people, more than of law professionals.

description model and language used to represent legal knowledge. In the 90’s, the development of knowledge-based processing systems promised to use technology to process the contents. Such development was hindered not so much by the intricacies of reproducing thinking schemes and processes pertaining to legislation, but by the lack of a knowledge-conceptualizing method. Nowadays, core ontology offers the basis for the development of ‘domain ontology’, knowledge basis describing specific normative sectors through definite, shared elements. This allows a rich gamut of advanced applications in knowledge analysis and management, such as: comparing and updating normative, assessing rule compliance, on-line advice and issue-settlement.

The JurWordNet database is accessible by itself, or integrated with the Italian database. The juridical sector will soon be available on this web site, both through Asp file connection, with information retrieval engines, and through navigation within the lexical net of ontology.