Periodic Reporting for period 2 - KNOWMAK (Knowledge in the making in the European society)
Reporting period: 2018-01-01 to 2019-12-31
The tool is based on three existing data sources on knowledge production in the European Research Area (ERA), i.e. scientific publications derived from the Web of Science database (CWTS publication database), patents derived from PATSTAT (RISIS patent database) and European projects derived from CORDIS (AIT-EUPRO database). Additionally, the project has developed an experimental dataset concerned with social innovation projects (European Social Innovation Database).
The tool was developed through a co-creation approach that involved a group of lead users early on in the project. The group of lead users was a core group of individuals who are archetypical for potential end user of the KNOWMAK tool in terms of their data/information need, their professional experience, and their affiliation to one of the target groups, i.e. policy-makers, research managers/funders, companies, regional actors, or civil society representatives.
The overall architecture of the KNOWMAK tool is depicted in figure 1. Primary data sources on knowledge production are structured and enriched in specific datasets, like the publication dataset maintained by the CWTS in Leiden; harmonization pertains to the three central integration dimensions in KNOWMAK, i.e. topics, actors and space. From these datasets, a core set of data are extracted and transferred to the KNOWMAK integrated database that feeds the construction of indicators and the on-line visualization tool.
The KNOWMAK project is a joint endeavor of eight partners with complementary expertise in the field of STI studies, i.e. the University of Paris Est Marne la Vallée (UPEM), the Austrian Institute of Technology (AIT), the University of Leiden (UL), the Politecnico of Milan (POLIMI), the University of Manchester (UMAN), the Zentrum für Soziale Innovation (ZSI), the University of Sheffield (UFSD) and the University of Strathclyde.
The project run from 2017 to 2019, the tool will be further maintained and developed within the Research Infrastructure for Research and Innovation Studies (RISIS2) Horizon 2020 project.
b) A consistent design of the KNOWMAK tool, including a selection of the indicators, has been constructed (see Deliverable D5.4). The selection of indicators has been based on the usage scenarios defined, so that the indicators provided by the tool will respond to the specific user needs. Further, indicators have been operationalized by the different levels of aggregation foreseen at the interface level. Composite indicators on knowledge production are also available.
c) The ontology is now fully operational and has been subsequently refined through different cycles of iteration and evaluation; the ontology was also enriched with additional keywords, and cleaned from additional stopwords in order to improve the identification of classes within data sources. The annotation service for data sources is also operational via a REST API that allows smooth annotation of the large sets of data available in KNWOMAK; the data tagged with KET/SGC topics have been transferred to KNOWMAK database in order to produce indicators (see Deliverables D2.4 and D2.5).
d) The existing KNOWMAK data sources have been enriched with standardized actors and with geographical information (see Deliverables D5.1 and D3.1). Public-sector actors have been harmonized by using the RISIS OrgReg facility, while work on the harmonization of firms (RISIS FirmReg facility) has been completed for large firms. Data sources have been geocoded and a NUTS3 adapted classification building on the EUROSTAT metropolitan areas has been adopted as regional classification in KNOWMAK. These results therefore allow displaying the KNOWMAK data by actor and by region across all geographical spaces.
e) The Data Management System is now fully operational and is populated with data. The indicators layer of the database has been implemented based on the list of indicators developed in WP5 (see Deliverable D6.4).
f) The database of social innovation projects is now released and the corresponding information is available in the KNOWMAK tool. Indicators on user attention (based on tweets) and on open access are also available in the tool.
g) Finally, the draft data management plan has been work out, that foresees a generally open approach to KNOWMAK data, and remaining ethical issues have been cleared. Dissemination activities have also been implemented and usage of the tool has reached the expected level.
The work performed on user needs has allowed to be more precise on how these impacts will be achieved and to identify usage scenarios on the tool linked to specific forms of impact:
• Examine national RTDI performance (policy-makers & research funders)
• Looking for evidence of impact of national research funding (policy-makers & research funders)
• Exploring emerging research areas and central actors (policy-makers & research funders)
• Using national funding data (policy-makers & research funders)
• Exploring the ERA performance in-depth (policy-makers & research funders)
• Identifying potential innovation partners (Research managers)
• Exploring public-private collaborations in emerging topics (Research managers)
• Looking into their regional performance, also compared to similar regions throughout the ERA (Regional actors
• Examining topical strengths a region and regional networks (regional actors)
• Identifying regional inequalities in RTDI (civil society)
• Exploring matching R&I partners (research managers)
At the technical level, the KNOWMAK project provided the following advances when compared with existing tools:
• The ability to combine on the same tool very different types of data, including publications, projects, patents and data on social innovation.
• A high level of standardization of knowledge production actors and regions.
• The adoption of ontologies as suitable bridges between policy questions and the research output.
• A user-centred design that puts users at the core of the processes and consequently builds a tool based on their needs.