The edge weight is the number of teams that attended both regionals & edges with a weight of 1 are ignored
Modularity: 0.584
Number of Communities: 14
The edge weight is the number of teams that attended both regionals & edges with a weight of 1 are ignored
This is pretty cool. I made something similar for the 2017 Ontario districts in January. There was some movement of teams after I generated the graphs, so the connectivity numbers won’t be perfect. I did two versions, one with labelled edges and one with weighted edges.
http://i.imgur.com/sMsFoKHl.png
Not gonna lie, it looks like you’re trying to summon a demon with these… :rolleyes:
I’m sort of surprised at the connection between San Diego and Las Vegas.
Thanks for the map.
If a team goes to 3 regionals, does it add a line from each regional connecting it as if 3 teams went to 2 regionals? (Sort of like a triangle)
How were communities derived?
Yes, but each line would only show up as long as one other team went to that pair of regionals (due to the edge weight restriction I added)
I generated the communities using the modularity statistic in Gephi which apparently uses an algorithm from this paper. I adjusted the resolution parameter to around 0.3 to get the number of communities I wanted.
Apparently not that algorithm, as it does not have a resolution parameter. Also, if I’m reading the algorithm and the graph correctly, the first pass would have moved the Hawaii Regional out of community with the Australian Regionals and into the SoCal/Las Vegas community. Thanks for the link; quite enlightening!