New EC-OECD International STI Policy Database

To be analysed and made openly accessible, country data needs to be stored in a database. The current database is publicly available (STIP Database). But the data collected remains under-utilised, largely because of outdated IT infrastructures that hamper accessibility for analysis. The project will use existing semantic database technologies that rely upon the STI policy taxonomy developed in WP3 to structure and link country data for analysis and open access. The main tasks in WP4 are to develop search and analysis interfaces that allow researchers and analysts to perform faceted search of country data in the database (e.g. by keyword, country, topic, pre-defined policy query, etc.), graphical navigation of the database contents, statistical analysis of the data and comparisons between countries, and extraction of reports that capture and display the results of research and analysis. While the database and its interfaces are targeted at an expert analyst-researcher audience, the project will pay particular attention to the user interface experience, including data visualisation and database navigation issues.

An important component of REITER concerns improving open access to the survey data for analysis. Much improved user interfaces for searching and analysing data that utilise visualisation tools, dashboards and standard queries will help unlock the data’s analysis potential and are expected to increase the data’s utilisation in future policy analysis. Accordingly, over the next few months, REITER will (i) deploy a new semantic database that provides enhanced data analysis capabilities and supports interoperability with other data sources; and (ii) develop improved search and analysis interfaces that make country information more accessible to analysts and researchers.

Country data needs to be stored in a database if it is to be analysed and made openly accessible. REITER uses semantic database technologies (MarkLogic) already deployed by the OECD in other contexts. These technologies provide much greater analytical flexibility than a traditional relational database and can admit any type of data. The database relies upon a so-called “taxonomy-ontology” of the STI policy field, which consists of concepts and the relations between them. This is used to structure and link country data for analysis and open access. MarkLogic supports both textual (Google-like) and semantic search. The latter supports data integration and more powerful querying, enabled by the STI policy taxonomy-ontology.

To enable users to access and use the database, REITER is developing user interfaces that will allow users to perform faceted search of country data (e.g. by keyword, country, topic, pre-defined policy query, etc.), graphical navigation of the database contents (e.g. using network maps), statistical analysis of the data and comparisons between countries (also supported by visualisation tools that enable charting and mapping of data), and extraction of reports that capture and display the results of research and analysis. These user interfaces will be developed gradually during the spring and early summer of 2017. It is essential that future users of the database, including country experts and delegates, are involved in this development process.

Content Visibility

Public