Hi Patrick thanks for the comment, not sure if you had a chance to read through the whole post, but it addresses all of your questions.
- The data came from Wikipedia textual data, it can be updated by manually reading the hundreds of pages that make up the country timelines. As much as I would like to keep it updated and I don’t have the resources. In retrospect, using tracking in the title may not have been the right word.
- I’m not sure what you mean by left censured? Wikipedia is open source and everything I’ve collected is from there. If what you mean is why I haven’t shared the data that i collated, all you have to do is ask nicely :) I did say i’d put it up on github if there was interest (with the obvious reliability disclaimer attached).
- there is absolutely no attempt at forecasting or prediction here. I made that crystal clear in the post. I have no domain knowledge in virology or epidemiology, indiscriminately throwing a ML algorithm at the data will give meaningless results. I’ll leave the forecasting to the experts. What this post was instead, was an exercise in data collation and visualization that attempted to do something remotely novel (because all i’ve seen is people throw the WHO data into a dashboard or fit exponential curves to cases) and try visualize the relationship of the spread that would have otherwise only be captured in textual data.
hope that helps,