studying covid-19 with machine learning — April 2nd 2020 — includes python notebook code — Part 1

Photo by Kay Lau on Unsplash

I have been using cover data for some time now, as a live project for my company, and to help my team build their machine learning skill.

Previous reports are available at the following links.

Focus — Regions

Alright, let’s begin with the current edition. In this edition, I am focusing on the following countries.

  • World Wide
  • United States
  • Sweden
  • Spain
  • South Korea
  • Italy
  • England

I chose the option above regions for some simple reasons. For one thing, I want to know what is happening with whole world. For instance, look at this plot below.

In the beginning, most of the data was driven by numbers that were driven by cases that were being reported in China (not included in today’s plot). So, there is an initial escalation of numbers , and then some flattening. Then, the numbers start picking up around March 1st week.

After March 1st week, things start getting really bad. This shows a number of things. For one thing, earlier, the virus was spreading in China. It was primarily spreading inside China. Then, as people started traveling all over the world, they took the virus with them. The virus has a 2 week period before it really starts affecting people. So, 2 people infect 2 people. Then, that becomes 4, and that becomes 8 and so on.

Further, there is also the delay with ramping up testing capacity. As the number of cases started rising, the number of people getting tested, and the capacity of a country to conduct tests, also started increasing. That is why after a relatively quiet period between February 15th to March 1st, things start picking up.

Now, we are seeing an exponential increase in numbers, which correlates with the exponential increasing in testing capability. There is no way to confirm this but the tests are only covering a portion of the actual infections, so the actual numbers are probably higher.

To Be Continued

One thing though, comparing the worldwide plot, with that of other countries, it is clear the world is still in the early stage of exponential case rise. It is at least 2 to 4 weeks before peaking (may be not even then, due to the nature of the virus).

As of now, we are very much in the aggressively growing stage of total infections. I will dig deeper into the projections for Worldwide as well as other regions in the next blog posts.

Stopping, for now.

I also have a podcast. You can listen to it at this link — Jay’s Podcast.

--

--

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Jay

Jay

Freelance Tutor — Coding, Generative Art and Photoshop. Full Name : Vijayasimha BR