Comparison of different clustering techniques for traffic noise analysis in the city of Milan

In this paper, we present the results of a comparative analysis on the traffic noise in the city of Milan within the framework of a co- financed project by the European Commission through the Life+ 2013 program called Dynamap. Dynamap is based on the idea of finding a suitable set of roads that display similar traffic noise behaviour (24 hours temporal noise profile). The dataset is made of the traffic noise recorded in 93 sites distributed over the entire city. Different unsupervised clustering algorithms have been studied: the so-called hard and soft clustering. In order to apply efficiently the Dynamap method, it is important that we have a continuity or soft passage between clusters as the non-acoustic parameter, which has been introduced to describe non-monitored roads, changes. The hard clustering coupled to a density distribution of the non-acoustic parameter in the two obtained road- cluster behaviours, revealed more efficient than the soft one though its inherent smooth approach to clustering.