際際滷

際際滷Share a Scribd company logo
Science maps and strategic
thinking
Flemming Skov, Tobias
Wang
Department of Bioscience,
Aarhus University
Jens Peter Andersen
Danish Centre for Studies on
Research and Research Policy,
Aarhus University
Mapping concept
 Topic representations based on co-keyword networks.
 Visualisation of aggregate units (authors, journals, institutions,
research groups) as clouds.
 Temporality.
 Directionality.
Strategic thinking
 Strategies for developing research portfolios:
 Stay (and fortify)
 Invade new areas
 Merge and fuse subjects
 Apply methods from other areas
 Explore whats beyond the border
 Invent new territority
The Atlas: Basemap for analysis
 Flexible approach.
 Source data: terms generated from abstracts, keywords,
descriptors etc.
 Create network of terms (nodes) and co-occurence in
documents (edges).
 Layout network as basemap, e.g. VosViewer, Kamada-Kawaii,
Force Atlas, OpenOrd.
 Basemap Cartesian coordinates as input for further analyses.
Cloud representations: fuzzy point
estimates
 How do we represent aggregate units, e.g. authors?
 Fixed point, average (x,y) of all contained keywords?
 Hull representation, covering all contained keywords?
 Density/fuzzy point
Case:
Zoophysiology at Aarhus University and
globally
 Data:
 Seed publications are those published by the section for Zoophysiology
at the Department of Bioscience at Aarhus University
 Additional publications from prominent journals and authors included.
 Adjusted approach: Related publications are extracted from Web of
Science by including other papers from the same microclusters
(Waltman & Van Eck).
 Approach:
 Co-keywords (descriptors and identifiers)
 Gephi for mapping, Force Atlas 2
 Map visualisations generated in Python
Atlas: Zoophysiology
Movement from physiology towards eco-
physiology
Keyword density
Three main areas
Author specificity
Author specificity 2
Three institutions
Keyword temporality
Where are the study organisms?
STI 2018 : science maps and strategic thinking
Strategic considerations
 Aarhus University has three strong research areas within
zoophysiology (traditional zoophysiology, physiology and
toxicology, audiology and marine mammals).
 Global attention shifts towards new areas (eco-physiology), where
Aarhus University is less represented.
 Should we invade? E.g. hiring a new professor, established in eco-
physiology; or focus attention on topics in that area.
 Should we stay and fortify? E.g. build further on established work.
 Can we learn something from other research areas, merging our
research with theirs? Are methods, topics, objects (e.g. study
organisms) possible to combine or align?
Thank you for your
attention
Twitter: @ipoga  slides will be available here soon.

More Related Content

STI 2018 : science maps and strategic thinking

  • 1. Science maps and strategic thinking Flemming Skov, Tobias Wang Department of Bioscience, Aarhus University Jens Peter Andersen Danish Centre for Studies on Research and Research Policy, Aarhus University
  • 2. Mapping concept Topic representations based on co-keyword networks. Visualisation of aggregate units (authors, journals, institutions, research groups) as clouds. Temporality. Directionality.
  • 3. Strategic thinking Strategies for developing research portfolios: Stay (and fortify) Invade new areas Merge and fuse subjects Apply methods from other areas Explore whats beyond the border Invent new territority
  • 4. The Atlas: Basemap for analysis Flexible approach. Source data: terms generated from abstracts, keywords, descriptors etc. Create network of terms (nodes) and co-occurence in documents (edges). Layout network as basemap, e.g. VosViewer, Kamada-Kawaii, Force Atlas, OpenOrd. Basemap Cartesian coordinates as input for further analyses.
  • 5. Cloud representations: fuzzy point estimates How do we represent aggregate units, e.g. authors? Fixed point, average (x,y) of all contained keywords? Hull representation, covering all contained keywords? Density/fuzzy point
  • 6. Case: Zoophysiology at Aarhus University and globally Data: Seed publications are those published by the section for Zoophysiology at the Department of Bioscience at Aarhus University Additional publications from prominent journals and authors included. Adjusted approach: Related publications are extracted from Web of Science by including other papers from the same microclusters (Waltman & Van Eck). Approach: Co-keywords (descriptors and identifiers) Gephi for mapping, Force Atlas 2 Map visualisations generated in Python
  • 8. Movement from physiology towards eco- physiology
  • 15. Where are the study organisms?
  • 17. Strategic considerations Aarhus University has three strong research areas within zoophysiology (traditional zoophysiology, physiology and toxicology, audiology and marine mammals). Global attention shifts towards new areas (eco-physiology), where Aarhus University is less represented. Should we invade? E.g. hiring a new professor, established in eco- physiology; or focus attention on topics in that area. Should we stay and fortify? E.g. build further on established work. Can we learn something from other research areas, merging our research with theirs? Are methods, topics, objects (e.g. study organisms) possible to combine or align?
  • 18. Thank you for your attention Twitter: @ipoga slides will be available here soon.