Tuesday, February 26, 2019

Mapping Countries by Similarity


The Grand Prize winner in the World Data Visualization Prize 2019 was an interactive visualization which groups the world's countries using artificial intelligence. This interactive visualization, created by Interacta, shows the connections and similarities between different countries based on a number of different factors.

The Country t-SNE map rearranges the countries of the world so that instead of being geographically positioned they are grouped by similarity. It uses the t-SNE machine learning algorithm that groups similar objects closer together and dissimilar objects further away.

A good way to start exploring the map is to click on the speech bubble icon at the bottom of the map. This opens a short introductory tour and explanation of the groupings that you see on the map. The countries on the map are grouped by their similarity in a number of different social and economic factors. If you click on an individual country's marker on the map you can see a list of these factors and how the selected country scores in each of these areas.

If you click on the gear icon, on the right-hand side of the map, you can adjust the different variables by which the countries are grouped on the interactive map. Adjusting these settings will regroup the countries on the map to show their closeness using the factors that you have chosen.

1 comment:

Anonymous said...

This is truly brilliant, well done and about time. AI could be of great benefit to the human race if we use it intelligently, tune it to respond to the most basic ethical compass and off we go.
Green Maiden