Calculating clusters with AI methods

Clever Fellow


A human observer can register clusters in a two-dimensional set of points at a glance. Artificial intelligence has a harder time getting it done; however, the relatively simple k-means method delivers usable results.

Nature lovers who tagged along with the previous edition of this column and generated a map with all US national parks might subsequently ask themselves how they can tour all these attractions using as few resources as possible. Figure 1 shows that the parks are concentrated in certain areas. A tourist can thus visit about a dozen spectacles of nature by focusing on one area during a single visit.