 In the previous video, we explored COVID-19 data and observed curves in time. But we also noticed our data is reported per country, so it would be even better to plot this information on an actual map. For the following example, you will need Orange's Geo and Time Series add-ons, which you can install from Options add-ons. Let us have another quick look at our data. We have information on the latitude and longitude of the country. How convenient! We'll use the GeoMap widget to plot data points. Latitude and longitude are automatically selected. What's even better, I can set the size of the point to reflect the number of cases per country for the selected day. We can see it starts big with China. Now I can continue pressing the down button and the size option will go through all the dates in the data, changing point sizes accordingly. Nothing changes for a long time, but around mid-March, China starts to get small and other points emerge. First Italy, Spain and Germany, then slowly the United States. This visualization is just a quick overview of the global pandemic. Observing many small points on a map is, however, very tiresome. We also mentioned that the data for some countries are split by territories, and it would be nice to show this data for the entire country. Core plus to the rescue. This widget aggregates data into regions, say provinces, states, countries. Select the latest date from the data and set the aggregation function to some. The colored regions now show the sum of confirmed COVID-19 cases for each country. But let's not forget, we're working with time series. What we're interested in are trends, how data changes in time. Let's animate our maps. We'll use time slice, which outputs a subset of the data for the selected timeframe. However, time slice requires a single time column with dates in rows, but we have dates in separate columns. First, I'll use select columns to move latitude and longitude to meta attributes to exclude them from transformations. Then I'll use transpose to turn columns into rows and vice versa. I want my column names to be countries, so I'll set country slash region in the from variable option. As always, the data table helps us check if we did everything right. Great, but we still have to tell orange to interpret the new column as a time variable. Connect edit domain, find the variable feature name, and change its type to time. We can also rename it to date while we're at it. The data is now ready for time slice. First, I'll select a single day of our COVID data. Don't press the play button just yet. We need another transpose, because we want to see countries on a map. And for this to happen, each country has to be in a separate row. This time, I'll use generic transpose. Let's see our data on a map. Connect coroplas to the second transpose. Select feature one as attribute and sum as aggregation. I'll hide the controls for a nicer view and place time slice next to the coroplas. Now it is finally time to press the play button. What we see is an interactive video of the pandemic. Time slice passes through our data a day at a time and sends it to coroplas, which then shows the sum of that day's reported cases per country. We were able to assemble a few widgets from time series and geo add-ons into an interactive workflow that enables us to observe changes in time. You can use the same workflow with GeoMap or other visualization widgets. For more details, see our blog, whose link is in the description below. And don't forget to stay tuned for more videos on analyzing COVID-19 data with Orange.