We present a case study of requirements for “data wrangling” capabilities in a healthcare application context. Data wrangling is an increasingly common requirement for data scientists, policy makers, market researchers, intelligence analysts, and other professions where existing data must be used in ways that were not envisioned when it was first collected. We characterise data wrangling as a programming problem, in which aggregate data must be restructured in ways that remain consistent with its semantic origins or ontological referents. We recommend the table as a lowest common denominator representational device, affording both direct manipulation and programming by example. We describe work in progress, in which we have identified new opportunities for clinical end-users to interact with the content of a customisable information system, through a focus on tables as an approachable analytic tool.
Type of Publication: Paper
Conference: PPIG 2016 - 27th Annual Conference
Publication Year: 2016
Paper #: 25