From Data to Knowledge
Collaborative session notes:
- IG ELIXIR Bridging Force
- IG Preservation Tools, Techniques, and Policies
- IG From Observational Data to Information
- WG Bio/FAIRsharing
Brief introduction describing the activities and scope of the group(s)
The scientific enterprise relies on knowledge infrastructures, environments that enable turning ever increasing amounts of data into knowledge. Knowledge infrastructures are underpinned by "robust networks of people, artifacts, and institutions that generate, share, and maintain specific knowledge about the human and natural worlds" [Edwards, P. (2010). A Vast Machine: Computer Models, Climate Data, and the Politics of Global Warming. ISBN: 9780262013925. MIT Press. http://knowledgeinfrastructures.org/]. As institutions of knowledge infrastructures, research infrastructures generate and handle research data of significant volumes while e-Infrastructures enable data-driven science. As artifacts of knowledge infrastructures, research data are knowledge-based resources [EU Regulation No 1291/2013, Title I, Article 2 (6). http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2013:347:0104... of interest to thousands of researchers across scientific disciplines. It is primarily researchers that turn the significant volumes of data handled by research infrastructures into knowledge to advance scientific understanding. Researchers, research data, research infrastructures, and e-Infrastructures are thus key elements of networks that generate, share, and maintain specific knowledge about the human and natural worlds. Connecting researchers and research communities to interoperable infrastructures is essential for data to evolve to knowledge.
This reality poses a number of challenges, such as:
- How to deeply integrate researchers and research communities with interoperable infrastructures?
- How to provide data-driven scientific workflows as a service by infrastructures to research communities?
- How can infrastructure support the flexible specification of workflows as required by research communities to support their scientific activities?
- How to efficiently utilise the large amounts of data now available?
- How to systematically acquire and curate data and their meaning (i.e. information) resulting from data analysis (e.g. data interpretation) activities carried out by research communities.
FAIRification of metadata and data plays an important role in this context. Indeed, the FAIR Principles will reduce the friction in using observational, experimental, computational (primary) data published by research infrastructures. However, from data to knowledge, the FAIR Principles and technologies (e.g. Linked Data Technologies) will need to be applied equally to data derived in analysis, i.e. data generated by individual researchers and research communities, and higher level data products (e.g. statistical data). Furthermore, down-chain analyses need to be findable, accessible, interoperable, and reusable. This includes workflows, in particular human (researcher) in-the-loop data analysis and interpretation, along value chains.
How to efficiently utilise the large amounts of data now available? “From Data to Knowledge”, i.e. from observational, experimental, computational primary data to information derived from data about human or natural worlds of interest.
“FAIRification” will help to make metadata (and hopefully data) FAIR, but in order to get useful information from data, also down-chain analyses need to be well documented. This includes workflows, in particular human (researcher) in-the-loop data analysis and interpretation, along value chains. Furthermore, interactions between the social infrastructure and technical infrastructure (including research infrastructure, e-Infrastructure, EOSC, etc.) matter.
- Understand the concept of knowledge infrastructure, such as the EOSC, and how it applies to modern research environments.
- Present ways to obtain information from data in various disciplines. Learn about techniques, e.g. to extract/represent/curate information, and knowledge.
- Discuss the relevance and role of the FAIR Principles to the evolution of information and knowledge from data in various scientific fields.
- Discuss approaches to deeply integrate researchers and research communities with interoperable infrastructures.
- Introduction to the session (Rob Hooft, Bengt Persson, 5 min)
- Knowledge infrastructures: Brief introduction (Markus Stocker, 10 min)
- Turning primary data into information: Concrete examples in various domains with presentations by From Observational Data to Information IG and Preservation Tools, Techniques, and Policies IG (Ruth Duerr et al., 15 min)
- Why FAIRification matters to evolving data into information and knowledge (Luiz Bonino, GO FAIR,15 min)
- The proposed FAIRmetrics (Susanna Sansone, 10 min)
- Discussion (35 min)
- Proposed topics reflecting the challenges listed above and objectives of the meeting, e.g. integrating researchers with interoperable infrastructures
- Topics raised by the audience
Additional links to informative material related to the participating groups
Everyone interested in how to facilitate to get information out of large scale data
Group chair serving as contact person: Bengt Persson
Type of meeting: Working meeting