Elemental diagrams

This tweet came through my feed today:

Twitter user @MeijaJuris noted that this image is a bit out of date, given that nihonium (element 113, Nh) shares its abbreviation with New Hampshire. I wanted to see how to use Mathematica to update the diagram.

Here’s the code:

(* Grab data - state and element abbreviations *)
elements = 
  EntityClass["Element", All][
   EntityProperty["Element", "AtomicSymbol"]];
states = Capitalize@ 
    EntityClass["AdministrativeDivision", "AllUSStatesPlusDC"][
     EntityProperty["AdministrativeDivision", "StateAbbreviation"]];
(* Create overlap group and 'leftovers' *)
int = Intersection[elements, states];
juststates = DeleteCases[states, Alternatives @@ int];
justelements = DeleteCases[elements, Alternatives @@ int];
(* Create two regions that will be combined *)
a = Disk[{-0.5, 0}];
b = Disk[{0.5, 0}];
subsets = Subsets[{a, b}, {1, 2}];
(* Create the regions *)
regions = 
     RegionDifference[BooleanRegion[And, #], 
      BooleanRegion[Or, Complement[{a, b, EmptyRegion[2]}, #]]]] & /@ 
(* Use WordCloud to create region labels *)
wc = MapThread[
   WordCloud[#1, #2, WordSpacings -> {25, 25}, 
     ColorFunction -> Black] &, {{juststates, justelements, int}, 
(* Put everything together *)
 PlotLabels -> Callout[wc, Center, Background -> None], 
 LabelingSize -> 207, AspectRatio -> 1/1.09, Frame -> None, 
 ImageSize -> 400, BoundaryStyle -> Directive[Thickness[0.01], White],
 Epilog -> {Text[Style["Elements", 14, Bold], {0.65, 1.1}], 
   Text[Style["States", 14, Bold], {-0.65, 1.1}]}]
Updated Venn diagram of states and elements.

The code is largely adapted from the documentation with some additions from the Wolfram Knowledge Base (or whatever it’s called nowadays). Once you learn the syntax (and that’s a huge hurdle), Entities contain a wealth of information that can be useful in any number of applications. The easiest way I’ve found to explore Entities is by using the natural language input feature to perform a search, which is accessed by starting a cell with =, and then converting the cell to input form to see how it was done:

Making sense of Mathematica’s natural language input. Click to enlarge.

Still, there’s the tricky part of making sure you know what is a string and what isn’t. That part requires a bit of trial and error. Perhaps there’s a simpler way to do this, but for now, it’s an approach that is working for me.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.