#	Pagina
attuale pagina	/open-h2020/projects/199931/results.html

Report

Teaser, summary, work performed and final results

Periodic Reporting for period 1 - MIREL (MIREL - MIning and REasoning with Legal texts)

Teaser

Summary

The project MIREL aims at bridging the gap between the communities working in Mining (i.e., Natural Language Processing and Information extraction on legal texts) and Reasoning (i.e., formal and computational models to reason on normative knowledge, towards applications for compliance checking and decision support), which has mostly worked in isolation.

The project was retained for funding with a very high overall score (97.2%), due to the importance of legal informatics for the societies in the Big Data era. Big Data has been identified as a key topic for H2020;
European legislation, estimated to be 170,000 pages long may be deemed by some to be a little bit too small to be considered Big Data per se. However, since legislation is at the basis of and regulates our everyday life and societies, many categories of Big Data (e.g., medical records in eHealth, financial data, etc.) must comply with and are thus highly dependent on specific norms. Matching and annotating Big Data with legislative information will produce even more and richer Big Data.

Thus, Legal Informatics is experiencing growth in activity, also at the commercial level.

Although current technologies provide valid solutions to help navigate legislation and retrieve information, the overall usefulness and effectiveness of the systems are limited due to their focus on terminological issues and information retrieval while disregarding the specific semantic aspects of law, in particular its logical structure in terms of constitutive and regulative rules, which allows legal reasoning.

For these reasons, MIREL was proposed to promote mobility and staff exchange in order to create an inter-continental inter-disciplinary consortium in Law and Artificial Intelligence areas including Natural Language Processing, Computational Ontologies, Argumentation, and Logic & Reasoning.

Work performed

\"The work in WP1 was oriented towards: carrying out transdisciplinary research and devising conceptual models for legal knowledge representation and reasoning, devising formal languages for representing norms, policies, and values in the law, and Logics for modelling the interpretation of legal provisions. The activities done, described with more in details in partB, have been carried out by UNIBO, Data61, UNLP, UNS, UL, ZJU, ROIS, Stanford Univ. and Cordoba Univ.

The mentioned partners: (1) developed a family of logical systems to describe a non-classical operator which models chains of obligations and permissions and compensatory deontic statements, (2) addressed quantification in non-normal deontic logic, (3) defined a general argumentation framework for modeling statutory interpretation, (4) addressed theoretical research on the relation between classical and multi-relational semantics, (5) worked towards the definition of a legal ontology for the upcoming general data protection regulation, (6) studied inference patterns and properties of detachment in normative systems, (7) worked on dynamics of legal normatives and on accommodating temporary norms within normative systems, and (8) worked on formal argumentation, aggregative Input/Output logic, and a goal-directed epistemic logic.

The work in WP2 was oriented towards: designing ontologies for normative knowledge, developing and extending existing NLP systems for mining both concept to be linked to the T-BOX and named entities in order to populate the A-BOX, building GOLD standard corpora for syntactic and semantic analysis. The activities done, described with more in details in partB, have been carried out by INRIA, Cordoba Univ., APIS EOOD, UNITO, UL, Nomotika SRL, and Stanford Univ.

The mentioned partners: (1) worked on new developments of the legal systems Licentia, EuroCases, and Menslegis, involving named entity recognition, ontology population, and text similarity in particular, (2) conducted research on named entity recognition, textual entailment, and question answering.

The work in WP3 was oriented towards: ontology-based access to normative knowledge, computational solutions for decision making and compliance, massive parallelization. The activities done, described with more in details in partB, have been carried out by HUD, DLVSystem, Data61, UL, UCT, UNS, ZJU, and UNIBO.
The mentioned partners: (1) worked on reasoning over legal rules and legal cases, as well as on massive parallelization, using Answer Set Programming, (2) implemented those methods within the system DLV, focusing on paracoherent (or paraconsistent) reasoning, (3) worked on argumentation, by modelling obligations and their compliance in Defeasible Logic, (4) worked on prioritized and non-monotonic reasoning in formal argumentation, (5) investigated novel parallelization techniques for inconsistency-tolerant query answering on ontological databases in the Datalog+/- language, (6) conducted research to integrate belief change and non-monotonic reasoning as well as on reasoning based on defeasible rules.

The work in WP4 was oriented towards: user needs study, licenses and contracts case study, technical documents case stud, and the multilingual corpora of norms case study. The activities done, described with more in details in partB, have been carried out by INRIA, Cordoba Univ., UNITO, Data61, and UL.

The mentioned partners: (1) conducted research on Ontology Population, in mining norms under the form of rules from the \"\"Australian Telecommunications Consumer Protections Codeâ€, (2) studied how semantic web frameworks could apply to the formalization, publication and processing of legal knowledge, (3) designed the questionnaire for identifying the users\' needs and conducted a marketing analysis of the legal informatics industry in the United States.\"

Final results

During the first Reporting Period, the work done in MIREL led to 63 publications, among which 15 journals, in top venues in legal informatics, greatly contributing to the progress beyond the state of the art.

We believe the MIREL succeeded so far in addressing the general social impact, stated in the beginning, for which it has been devised, i.e., bridging the gap in legal informatics between the community working on mining and the one working on reasoning. This has been achieved via the organization of several conferences, workshops, and events specifically designed to this end. As a result, new researchers and companies not originally belonging to the consortium are currently working with MIREL participants, e.g., Nuance San Francisco and Freie University of Berlin.

Moreover, the coordinator University of Luxemboug decided to create a portal called â€œLegal Informatics Luxembourgâ€ gathering together key players in Luxembourg, Europe, and worldwide, and to list the R&D activities, events, and collaborations currently carried out by the members of the community, in order to continue the dissemination, communication, and networking strategy of MIREL beyond the ending date of the project. The ambitious goal of the portal is to be an hub for all key players in legal informatics, and a reference website for whoever is willing to start working on the field. A demo of the portal is currently under construction and available at http://www.luxli.lu/.