O Data Catalog Vocabulary (DCAT) para a publicação de dados de pesquisa nos princípios Linked Data

Abstract

With the development of semantic technologies, the importance and the recognition of data as information sources grow. In this context, the data generated from researches are increasingly valued, since their publication allows the foundation of the results obtained with the scientific works, also helping in the production of new knowledge in a sustainable way. Thus, there is a need to organize these research data, so as to optimize their retrieval. For this purpose, vocabularies could be used. From the standardization of data representation, vocabularies allow data to be organized and retrieved efficiently. When considering the Web environment, the discovery of new data is also improved, using Linked Data principles, and the Data Catalog Vocabulary (DCAT) appears as a vocabulary aimed at describing catalogs and its datasets, seeking to favor their interoperability. The objective of the research is to analyze the characteristcs and fundamentals of the Data Catalog Vocabulary (DCAT) and the vocabularies adopted by it, for the publication of research data in the Linked Data principles, in order to identify the discussion on the subject in the literature and verify the use of it. Through exploratory, descriptive and theoretical research, it was possible to systematize and analyze the characteristics of the vocabularies, identifying that DCAT allows catalogs and datasets to be standardized, reducing ambiguity issues and providing the discovery of data and data services, since it has specific terms for this purpose. Thus, in line with Linked Data principles, the use of DCAT allows the benefits of sharing research data to be accomplished. An analysis of re3data platforms was also carried out in order to identify the current use of vocabulary by them, where we verified low adoption of DCAT. However, the results indicate the concern of these platforms with issues of findability, accessibility, interoperability and reuse, going against the Linked Data principles and the FAIR principles. The research revealed that there are few discussions about DCAT in the literature, especially in the scope of Information Science. Thus, we point to the need to better explore the possibilities of representing information in the context of research data. In addition, further discussions that align the knowledge of the area with the semantic tools and concepts applied to the Web enhance the benefits of finding, using and retrieving information, in addition to expanding and improving the knowledge and possibilities of information professionals' performance.