COVID-19 Live Report
Authors: Yutong Song, Behrooz Hassani-Mahmooei
Working with Shiny < 1 year
Abstract: Our submission to Shiny Contest 2020 is triggered by COVID-19 affecting daily lives globally, we aimed develop this app to visualise the trends and make them available to everyone.
Full Description: Objective
This app was developed to make the data and key visualisations of COVID-19 trends available to everyone and also provide a platform for further detailed analysis of the trends.
Methodology
The data is sourced from the code developed by Tim Churches (UNSW) which extracts data from Johns Hopkins University . In addition, we sourced data about country and province/state population from Wikipedia.
This application differs from other COVID-19 applications by focusing on number of days since the first cases report from each selected country instead of calendar dates. Using this application, countries can be compared easily based on their actual and per capita performance as the Y-axis scale is logged.
In addition, the application allows linear regression and smoothing patterns for selected countries, and compare the growth rate with a fixed growth rate.
Lastly, the application integrates World Bank data and policy data (thanks to Oxford University) to form country profiles for its capacity and readiness.
Main Packages
We used shiny and shinyWidgets for the mainframe of the application, used tidyverse, and lubridate to prepare data, and applied scales, ggrepel to improve the readability of visualisations.
Limitations
Data source may contain information that we could not explain. For example, Japan had a minor drop in their cumulative trend between 2020-01-22 and 2020-01-23. Also, the number of cases is highly driven by the testing practices across countries so we focused on confirmed cases at the moment, however death and recovery trends are also added.
Contribution
We aware that teams at the World Bank and Red Cross, and few other organisations are following the app updates to inform their COVID-19 impact analysis.
Acknowledgements
We acknowledge and appreciate the support that the RStudio team provided by offering an unlimited access account for this application, and Time Churches (UNSW) and Johns Hopkins University (JHU) for data sharing.
Thanks for Viewing our APP.
Sincerely,
Yutong, Behrooz
Please feel free to contact us via
Behrooz Hassani-Mahmooei Behrooz Hassani-Mahmooei, PhD - Monash University | LinkedIn
Yutong Song https://www.linkedin.com/in/ytsong/
Twitter
Behrooz Hassani-Mahmooei @behrooz_hm
Yutong Song @yurisyt
Category: Public Sector
Keywords: coronavirus, trends, time-series, analysis
Shiny app: https://behroozh.shinyapps.io/COVID19/
Repo: GitHub - yytsong/COVID_19: behrooz and I developed this app for analysing trend of COVID-19.
RStudio Cloud: Posit Cloud
Thumbnail:
Full image:
