Automated data collection from bitmap images on Polish Health Ministry's Twitter. Some thoughts on COVID-19 statistics in Poland.

Speaker(s)

Anna Ochab-Marcinek

Affiliation

Institute of Physical Chemistry, Polish Academy of Sciences

Date

May 27, 2020, 12:15 p.m.

Information about the event

meet.google.com/ufe-xfwd-jio

Seminar

Seminar of Biomathematics and Game Theory Group

---

I will only briefly focus on the technicalities. Polish Ministry of Health publishes a large part of the coronavirus-related data only in the form of bitmap images on Twitter. I wrote a set of Python scripts which use image filtering and OCR to automatically collect the numbers from the images. I publish these data in a numeric format on Github (anuszka/COVID-19-MZ_GOV_PL) and I visualize them on the website http://soft.ichf.edu.pl/ochab/coronavirus_poland. The main aim of this presentation is to share some thoughts on the statistics that I have visualized, possibly as an inspiration for mathematicians. I don’t do any modeling. I will only present a few observations that can be made when comparing the available data on confirmed cases, fatalities, hospitalizations and numbers of tests. I will also mention some other, not always widely known, Polish coronavirus-related projects or analyses that can be found on the internet. *** Hangouts Meet: at 12:00 *** meet.google.com/ufe-xfwd-jio

---