A stakeholder needs analysis (SNA) was employed in order to identify insight needs and use cases for using indicators and visualizations of emergence (and indirectly convergence) in daily decision making environments.
A user study was run to examine visualizations in terns of utility and readability.
-
User Needs Analysis
- online survey
- survey form: /User-Study/User Need Analysis Instrument.pdf
-
User Study
- online survey link
- survey form: /User-Study/User Study Analysis Instrument.pdf
Data Statistics: /code/dataset-crane-stats.ipynb
i. Web of Science database:
We extracted data consisting of two columns, WOS_id and WOS_ref from Web of Science database. Each row represents the source and target of references, via a WOS_id and WOS_ref unique identification number. We extracted three csv files, corresponding to each field of interest (AI, IoT and Robotics). See WoS relational database schema - Link: https://cadre.iu.edu/resources/documentation/wos_core_erd%20(1).pdf
ii. Web of Science Online Portal:
We used Institute of Scientific Information (ISI) publication data from Web of Science Online Portal to get the year and other relevant descriptions of papers. ISI files were converted to CSV using Make-a-Vis available at https://make-a-vis.netlify.com/
iii. NSF Award Portal
We used NSF award portal to extract active NSF award for the period of 1998-2017 - https://www.nsf.gov/awardsearch/
Using results from a linguistic algorithm comparison detailed in Börner et al. (2018), the MaxMatch algorithm (Wong & Chan, 1996) was used to identify terms in NSF funding awards that match the unique WOS Author Keywords specific to the three topic areas. Characters like [ ] { } and” were removed. See Appendix for extracted keywords disambiguation and normalization.
Maxmatch Algorithm is describe in the repository for the paper "Skill Discrepancies Between Research, Education, and Jobs Reveal the Critical Need to Supply Soft Skills for the Data Economy" https://github.com/cns-iu/cjobs.
Gephi (Version 0.9.2) (Bastian, Heymann, & Jacomy, 2009) was used to compute data overlays with the following plugins:
- ForceAtlas2 was used to layout the Co-Author Network figure
- GeoLayout was used to create 1) the Co-Author network overlaid on US map with mercator basemap and the Temporal Convergence figure using x-y coordinates to display nodes.
Data processing is described in Workflow CRANE (see Appendix folder).
Burst Detection uses Kleinberg's algorithm available in Sci2 (Sci2 Team, 2009). See the burst visualization code in Code directory.
The Convergence Network is described in Crane Workflow (see Appendix) and data convertion from WoS ids and WoS references into a 2D network is performed using the convergence code (see code directory).