Semantic processing of legal texts

The paper provides an overview of the field of semantic processing of legal texts, combining views and perspectives from the computational linguistic and Artificial Intelligence and Law (AI & Law) communities. The last few years have seen a growing body of research and practice in the field of AI & Law which addresses a range of topics: semantic and cross-language legal Information Retrieval, document classification, legal drafting, legal knowledge extraction, automated legal argumentation, as well as the construction of legal ontologies and their application. The increasing availability of legal corpora accessible as processable data is making viable their partially automated conversion into legal knowledge bases. In this context, it is of paramount importance the use of Natural Language Processing (NLP) techniques and tools that automate the process of knowledge extraction from legal texts. Accordingly, the paper aims at discussing how the two research communities can benefit from the interaction of the different perspectives: the legal artificial intelligence community can gain insight into state-of-the-art linguistic technologies, tools and resources, and the computational linguists can take advantage of the large and often multilingual legal resources (corpora as well as lexicons and ontologies) for training, domain adaptation and evaluation of current NLP technologies and tools. The authors will present an overview on semantic resources for legal texts annotation and processing. Different kind of resources (linguistic, lexical, conceptual, formal) will be introduced and their differences, methodological premises, intended use and possible integration will be highlighted. The peculiarities of the legal domain and legal language will be discussed in relation with the construction and use of legal semantic resources. The issue of multilingualism, multilingual and multi-legal system access to legal information will be also discussed showing how formalized lexical, linguistic and conceptual legal resources can support the task. How NLP tools and techniques can be fruitfully exploited to semantically process collections of legal texts will be introduced in the second part of the paper. In particular, the authors will show how they can be used to automatically extract the relevant knowledge contained in legal text corpora, to structure the extracted knowledge in semantic resources (such as domain-specific ontologies or thesauri), and to semantically annotate the texts with the extracted information to pave the way to content-based access and querying.