SEARCHING MADE EASIER
The JRC Dashboard on European data spaces can showcase the publicly available research by the JRC on topics related to data sharing and data spaces, allowing users to retrieve results using different keywords, without having to perform single queries on the JRC Publications Repository.
The dashboard is hosted on the JRC Big Data Platform (BDAP) and was created through the Voilà library. The code that performs the search was written in Python and at the moment the tool searches for publications on a dump of the publicly available JRC publications. In the near future we plan to perform this search on the live database, avoiding the need for periodic updates on the dump and including also the most recent publications on these topics.
HOW TO NAVIGATE
The homepage: the dashboard shows the main functional requirements as well as their descriptions, on different buttons that can be selected. Once a requirement has been chosen, three different blocks appear on the page.
On the left side: a list of keywords that can be used to retrieve relevant JRC publications. Two levels of keywords appear: the first one is composed by an expert-determined combination of words to set the search environment, while the second one constitutes a list of keywords similar to the first level and is automatically retrieved using the SeTA tool. All the keywords are by default already selected, and the user can decide to deselect one or more of them, updating the query through the “Retrieve publications” button.
The central block: composed of the list of publicly available JRC publications that were retrieved through the keywords. The page only shows the title and the year of the publication, but by clicking on each line the user will be redirected to the corresponding page on the JRC Publications Repository,23 where complete information is available, as well as the publication itself. The list of publications is downloadable in CSV format through the down arrow button.
The top right: panel includes filters that can be applied to the search. By default, the search of the left panel keywords is performed using the title, abstract and keywords of the publications, but through this functionality the user can decide whether to exclude some of these. The slide bar allows the user to expand or reduce the search with respect to year of publication. The bottom right panel includes a bubble chart that gives the user an indication as to the amount of publications retrieved by year.
Keywords: how the set of keywords - the aim is to create a mix of human-validated terms and machine-identified ones. In particular, starting from an expert curated list of keywords that represents the first level (taking terms from the title of the requirements and fine-tuning the list with some other concepts that could relate well to those topics), we then added to this list a series of similar terms retrieved through a semantic text analysis tool, SeTA (second level). We explicitly decided not to allow users to search open text keywords as this could lead to the retrieval of less relevant publications that do not actually relate to the realm of the requirements that were identified for common European data spaces.
Suggested Section: JRC Science for Policy Report or JRC Catalogue
Read the JRC Science for Policy Report or search the JRC Catalogue for JRC publications which support data spaces