KU Leuven/LIBIS announced the publication of a report on the data architecture for the ReIReS’ search platform, an important step forward in the development of the final platform.
The growing amount of data out there confronts researchers today with a new challenge, because it is spread out over different databases in different institutions. Therefore, ReIReS is developing a search platform for religious studies, which will collect relevant datasets on a single platform.
Before working on the data architecture, KU Leuven and Brepols drafted a Requirements Overview and Use Case Models (Deliverable D6.1), as well as an Integrated Data Model (Deliverable D6.3).
Designing the Data Architecture
The first step in designing the data architecture was the development of the back-end. This required solutions for the delivery, storage and retrieval of data. Datasets from different sources need to be delivered and stored in a uniform way. To that purpose, different methods of data delivery are planned for implementation and a system to validate delivered data has been developed. The data is stored in an AgensGraph graph database, which facilitates the expression of connections and relationships between data from different sources.
The aggregated data stored on the graph database is searchable together with datasets that are accessible through API, such as Index Religiosus, without those datasets having to be imported. Its publisher, Brepols, is developing this API for Index Religiosus. To ensure that the search results from both the graph database and the federated search via API are organized in a uniform result list, a blender has been implemented that merges the query results in a sortable list. This allows other databases to be included in the unified discovery environment in the future, provided they are accessible via API and map their data to the ReIReS data model. An Elasticsearch engine handles search requests in the graph database.
What’s Next
The next steps in the architecture design phase will focus on updates of this first version of the unified discovery environment and further integration work.
- A first release of the user interface is planned for early 2020.
- New features will be added to the user interface after testing and evaluation with users.
- Further configuration and refinement of the implemented components.
- More datasets will be added in the coming months and throughout the project.
Data Architecture Report
Go here to download the report (Deliverable D6.3).