The work carried out for this thesis focused on the implementation of extensions for a state-of-the-art clustering algorithm called Brige Aware Clustering. The aim of clustering is to analyze the data given as input and find similarities and differences among the data points. Data points are dived in groups, named clusters, in such a way that each point is similar to the points belonging to his cluster and dissimilar to points belonging to other clusters. The extensions for the Bridge Aware Clustering algorithm were made in two different directions: making the algorithm noise-tolerant, in order to be able to correctly classify dataset containing noise points, and implementing a distributed version of the algorithm, allowing the execution on large datasets.