Analysis of the evolution of the coronavirus

Inglés

In this post I want to share an analysis that I have done on the evolution of the coronavirus using different tools for data analysis and visualization: Python to obtain historical data from all affected countries: https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_19-covid-Confirmed.csv https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_19-covid-Deaths.csv https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/time_series_19-covid-Recovered.csv

KNIME for data manipulation and the creation of a predictive model using Random Forest to analyze the evolution of contagion before showing symptoms:

Facebook Prophet to estimate the behavior of the disease in the next 7 days.

Tableau for displaying the results and publishing the dashboard on your public server where you will find:

Case evolution: Where the evolution of the total cases of the disease from the origin is shown, as well as an adjustment in which deaths and discharges are subtracted. 7 days projected using the Facebook Prophet library are included.

Case incubation: Cases reported vs. probable cases before they can be identified. The model has been built using behavior in all affected countries using the Random Forest algorithm.

Death vs Recover: Cases discharged compared to deaths. A projection is also made for the next 7 days.

Case increment: This graph shows the unique cases identified day by day so that we can identify the spikes or stagnation of the disease depending on the country.

Here is a direct link to the dashboard on the Tableau public server:
https://public.tableau.com/views/coronavirus_analysis/Dashboard1?:display_count=y&publish=yes&:origin=viz_share_link

Este sitio usa Akismet para reducir el spam. Aprende cómo se procesan los datos de tus comentarios.