GOOGLE DATASET SEARCH: Overview and perspectives for indexing and availability of open scientific datasets

Abstract

In order to collaborate with scientific production in the field of data science, specifically in tools for storage and retrieval of data sets over the internet, this article aims to obtain an overview of the functioning, standards and perspectives on the Google Dataset Search tool - launched in 2018 with the proposal of identifying, indexing and making available internet datasets (massive sets of data) - essential instruments for the scientific community. The methodology used was descriptive, exploratory and bibliographic. A bibliographic survey was carried out on the platform, identifying internal functioning, standards, guidelines, formats and standardization institutions that guide the platform, in addition to current statistics of indexed data. Then, practical tests of use, usability and operation of the tool were performed, according to available documentation. The results obtained showed a promising platform, with a satisfactory usability score, aligned with international data interoperability standards and with considerable volumes of datasets already available, mostly in the English language. It was also observed, after the tests, that there are already several brazilian data repositories indexed by Google Dataset Search. However, some of them, even adopting the same metadata standards as this tool, are not yet available. The conclusion is that it is a system created by Google, with a high capacity for tracking, identification, indexing, interoperation and making available data sets available on the internet using international standards and, therefore, has significant potential. This work contributes to the large area that is inserted, reducing the scarcity of scientific publications on tools for making data sets available, specifically on the functioning, protocols, mechanisms and interface of this current tool.