Patricia Martín-Chozas

This paper describes an approach to represent terminologies in the machine-readable format of the Semantic Web, which improves the interoperability between terminological resources and opens up new possibilities yet to be discovered. The study’s motivation stems from the realization that the existing formalisms, such as SKOS or OntoLex-lemon, might not adequately capture the information within authoritative terminological resources. Therefore, we identified model requirements by formulating a set of Competency Questions derived from the analysis of terminological resources across various fields and domains, in line with the ontology development methodologies adopted in this work. During this analysis, we faced different representation challenges such as the various sources of term descriptions and the quality indicators related to terms. Consequently, we propose Termlex, a proposal based on the OntoLex-lemon model that combines the conceptual structure of the SKOS model with the lexical information as modelled in OntoLex-lemon. In Termlex, we define new classes and properties to cover the specific needs of terminological resources coming from a variety of approaches. The paper concludes with the instantiation of the Termlex model through three different use cases that follow different modelling approaches as a validation attempt.

With the rise of Artificial Intelligence driven services and Natural Language Processing techniques, high quality terminological resources are more needed than ever. Still, substantial manual effort is involved in the generation of such resources and many of them are published in non-machine-readable formats, hindering their reuse. In this chapter, we analyse how the Semantic Web can help represent legal language resources and how its publication in open formats can contribute to its interoperability. Throughout the sections, we give an overview of the Semantic Web, review existing legal language resources and the semantic models used to represent them. We also offer a series of practical examples to show the advantages, encouraging users to adopt these conventions.

Domain-specific terminologies play a central role in many language technology solutions. Substantial manual effort is still involved in the creation of such resources, and many of them are published in proprietary formats that cannot be easily reused in other applications. Automatic term extraction tools help alleviate this cumbersome task. However, their results are usually in the form of plain lists of terms or as unstructured data with limited linguistic information. Initiatives such as the Linguistic Linked Open Data cloud (LLOD) foster the publication of language resources in open structured formats, specifically RDF, and their linking to other resources on the Web of Data. In order to leverage the wealth of linguistic data in the LLOD and speed up the creation of linked terminological resources, we propose TermitUp, a service that generates enriched domain specific terminologies directly from corpora, and publishes them in open and structured formats. TermitUp is composed of five modules performing terminology extraction, terminology post-processing, terminology enrichment, term relation validation and RDF publication. As part of the pipeline implemented by this service, existing resources in the LLOD are linked with the resulting terminologies, contributing in this way to the population of the LLOD cloud. TermitUp has been used in the framework of European projects tackling different fields, such as the legal domain, with promising results. Different alternatives on how to model enriched terminologies are considered and good practices illustrated with examples are proposed.