Project for the 'Data Mining and Machine Learning' course. Please, check the Presentation for a complete view.
This is an implementation of a DBSCAN in a federated learning way.
The idea is to train a model, in a "collaborative" way, without transferring the data to a central processing server.
The algorithm was tested with various parameter settings and against different data sets. Its performance was then compared with that of a standard DBSCAN. No substantial difference emerged from this comparison.