Overall Objectives
Research Program
Application Domains
New Software and Platforms
New Results
Bilateral Contracts and Grants with Industry
Partnerships and Cooperations
XML PDF e-pub
PDF e-Pub

Section: New Results

Argumentation Theory

Combining Argumentation Theory and Normative Reasoning with Natural Language Processing

Participants : Serena Villata, Elena Cabrio, Fabien Gandon.

We have proposed a methodology to identify and classify the semantic relations holding among the possible different answers obtained for a certain query on DBpedia language specific chapters. The goal is to reconcile information provided by language specific DBpedia chapters to obtain a consistent results set. The results of this research have been published at the LREC conference [29] . This classification has then been exploited in another work, together with Elena Cabrio and Alessio Palmero Aprosio (FBK Trento, Italy), where Serena Villata has worked on an extension of QAKiS, the system for open domain Question Answering over linked data, that allows to query DBpedia multilingual chapters. Such chapters can contain different information with respect to the English version, e.g. they provide more specificity on certain topics, or fill information gaps. In particular, she extended the results presented last year embedding the new identified relations among the different answers, using argumentation theory to reconcile the information and further improving the system's performances. A demo of the new argumentation module is available online ( ). This work has also been presented at the International Semantic Web Conference demo session [85] .

Moreover, we have proposed, together with Alessio Palmero Aprosio, a system called NLL2RDF to translate in an automated way licenses, such as GPL, in natural language into a machine-readable version using the RDF language. The system is available online ( ). The results of this research have been presented at the European Semantic Web conference [26] .

Finally, we have published the benchmark of natural language arguments called NoDE. The benchmark is available online ( ). The results of this research have been presented at the 15th International Workshop on Non-Monotonic Reasoning (NMR 2014) [28] and at the 5th Conference on Computational Argumentation [27] (COMMA 2014 - demo).

Argumentation and Legal Reasoning

Participant : Serena Villata.

Together with Leendert van der Torre (University of Luxembourg), we proposed a framework for reasoning about norms using argumentation theory. Norms regulate our everyday life, and are used to assess the conformance of our behavior with respect to the regulations holding in specific contexts. Given the profound importance of norms in our lives, it is fundamental to understand which norms are valid in certain environments, how to interpret them, the legal conclusions of such norms, which norms can be derived from the existing ones, etc. In order to understand norms, people discuss about them to assess the validity or applicability of a certain norm subject to particular conditions, to derive the obligations and permissions to be enforced, or claim that a certain normative conclusion cannot be derived from the existing regulations. Several frameworks have been proposed for legal argumentation, but no comprehensive formal model of legal reasoning from arguments has been proposed yet. The goal of this work is to enrich legal argumentation with a formal account of deontic modalities. These results have been published at the 5th Conference on Computational Argumentation [55] (COMMA 2014).

Moreover, together with Guido Boella (University of Torino, Italy), Pietro Baroni and Massimiliano Giacomin (University of Brescia, Italy), Federico Cerutti (University of Aberdeen, UK), Leendert van der Torre (University of Luxembourg), we have studied also the dynamics of argumentation framework and this research has lead to a publication in the Artificial Intelligence journal [15] .