Edge-clustering

Implementation of a clustering algorithm using an agregation criterion[1] where clusters are formed of connected components with the smallest edges possible, that is, all of them are minimum spanning trees.

The hierarchy of modules is as follows

clustering(.py|.ipynb)
|---connected_components.py
|---prim.py
|---kruskall.py
    |---union_find.py

All the code is avaliable .py modules and as notebooks. The notebooks will be "recompiled" into .py modules when running the 'clustering' notebook in jupyter. That notebook contains a demo run and visualizations.

The console use is throgh stdin/out to facilitate piping:

$python(3) clustering.py [ ( kruskal | prim ) [ desired_number_of_clusters ] ] < data.txt > clusters.out

Note that this aggregation strategy has evident shortcommings wich are discussed alongside performance analysis in the repport (in portuguese)

[1] Gionis, A., H. Mannila, and P. Tsaparas, Clustering aggregation. ACM Transactions on Knowledge Discovery from Data (TKDD), 2007. 1(1): p. 1-30.