Where do data-driven insights come from? At Jumping Rivers, we believe that data should be made as interactive as possible, so that it can be used to support decisions throughout an organisation. Dashboards and other data-driven applications provide an accessible means for users to explore datasets, ask questions and make decisions, without those users needing the technical skills involved in processing and presenting data.
Here we present a gallery of data-driven applications developed at Jumping Rivers. We have a wealth of experience developing and maintaining data applications, using a variety of tools: Shiny (in both R and Python), Streamlit, Dash and a range of JavaScript libraries.
Litmus R Package Validation Dashboard
Shiny, R, Litmus, validation
A Shiny dashboard displaying quality- and risk-assessment scores for R
packages. Package assessments were performed using the public-version of
our Litmus analysis pipeline. Interactive graphics were created with
{echarts4r}, tables using {reactable} and the responsive design was
specified using CSS-grid through the {imola} package.
A Vue.js application to display flight departures, styled to look like the departure
screens found in airports. The user can interactively select which columns of data to display
in the departure screen, and can filter by date and time of departure. The departure screen
could be included in an application.
A Shiny-based map application that displays the boundaries for the "Integrated Care Board"
regions of NHS England. The leaflet-based map is interactive. On clicking a region, the user
can view statistics (population size etc) about that region.
A D3.js-based vertical timeline highlighting key moments in the development of the R language.
This was originally presented in our blog post celebrating the 30th anniversary of R.
Diffify: Compare different versions of R and Python packages
R, Python, Javascript
Diffify is a web application that provides a comparison between different versions of R packages
stored on CRAN and Python packages stored on PyPI. It is based on a Python back-end for analysing
differences between packages, and a React front-end for data presentation.
