This is a lot of fancy math and graphics that don't differ all that much from what a very simple analysis can give you. Only Revere belonged to five of the groups of interest. Of the three who belonged to four groups, two of them are in "top scorers" on centrality, and the third, with one of the two in centrality, is in the final table.
It's an exceedingly simple example used to illustrate the technique. The methods included in the python library are robust enough to work just as well on much larger and more complicated data-sets but that would be too difficult to follow for someone who isn't familiar with the subject.
I think it's hard to understand what the point is without a result that differs from the naive solution. That is, my naive "who's an important terrorist" is "who belongs to the most terrorist groups".
While belonging to the most terrorist groups will usually give the highest score, it's not quite the same as what betweenness actually measures - belonging to the most terrorist groups that are each large but do not share many members.
