We show that fastgreedy and walktrap community detection methods on weighted label co-occurence graphs are 85-92% more likely to yield better F1 scores than random partitioning.

Rona! Corona Mitchell! Corona Margarita! Cuckoo! Where've you gone to? Coo—ee! came in reply from the distance, and presently Rona appeared beckoning vigorously.

He that walketh in the darke, wotteth not whither he goeth.

The fire was restoked and the army of wine-bottles gave way to a smaller phalanx of brandies, Armagnacs and Marcs, to offset the large bowls of coffee from which rose plumes of fragrance.