Populating Legal Ontologies using Semantic Role Labeling

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 2157–2166 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 2157 Populating Legal Ontologies using Semantic Role Labeling Llio Humphreys1, Guido Boella1,5, Luigi Di Caro1,5, Livio Robaldo2,5, Leendert van der Torre2, Sepideh Ghanavati3, Robert Muthuri4 University of Turin1, University of Luxembourg2, University of Maine3, Strathmore University4, Nomotika SRL5 {humphrey, guido, dicaro}@di.unito.it, {livio.robaldo, leon.vandertorre}@uni.lu, rmuthuri@strathmore.edu, sepideh.ghanavati@maine.edu Abstract This paper is concerned with the goal of maintaining legal information and compliance systems: the ‘resource consumption bottleneck’ of creating semantic technologies manually. The use of automated information extraction techniques could significantly reduce this bottleneck. The research question of this paper is: How to address the resource bottleneck problem of creating specialist knowledge management systems? In particular, how to semi-automate the extraction of norms and their elements to populate legal ontologies? This paper shows that the acquisition paradox can be addressed by combining state-of-the-art general-purpose NLP modules with pre- and post-processing using rules based on domain knowledge. It describes a Semantic Role Labeling based information extraction system to extract norms from legislation and represent them as structured norms in legal ontologies. The output is intended to help make laws more accessible, understandable, and searchable in legal document management systems such as Eunomos (Boella et al., 2016). Keywords: Semantic Role Labeling, Legal Ontology, LegalTech 1. Introduction This article is concerned with a major obstacle to the goal of maintaining legal information and compliance systems: the ‘resource consumption bottleneck’ (Hepp, 2007) of cre- ating semantic technologies manually. Information extrac- tion techniques could significantly reduce this bottleneck. However, it has been argued (Lenci et al., 2009) that: Technologies in the area of knowledge manage- ment and information access are confronted with a typical acquisition paradox. As knowledge is mostly conveyed through text, content access requires understanding the linguistic structures representing content in text at a level of consid- erable detail. In turn, processing linguistic struc- tures at the depth needed for content understand- ing presupposes that a considerable amount of domain knowledge is already in place. The research question of this article is: How to address the resource bottleneck problem of creating specialist knowl- edge management systems? In particular, how to semi- automate the extraction of definitions, norms and their ele- ments to populate legal ontologies? This article shows that the acquisition paradox can be ad- dressed by combining state-of-the-art general-purpose NLP modules with pre- and post-processing using rules based on domain knowledge. The output is intended to help make laws more accessible, understandable and searchable in a legal document management system. Research on ontology learning (creating new ontologies) and ontology population (populating existing ones) is an important field of ontology engineering, albeit not with- out limitations: “none of the methods used today are good enough for creating semantic resources of any kind in a completely unsupervised fashion, albeit automatic methods can facilitate manual construction to a large extent” (Bie- mann, 2005). While ontologies can be learned from struc- tured …

Content truncated to 500 words