The idea underlying modal clustering is to associate groups with the regions
around the modes of the probability density function underlying the data. This
correspondence between clusters and dense regions in the sample space is here exploited
to discuss a possible extension of modal clustering to the analysis of social
networks. Such extension, albeit non trivial, seems particularly appealing: conceptually,
the notion of high-density cluster fits well the one of cluster in a network,
where groups are usually regarded as collections of individuals with dense local
ties in their neighborhood. Additionally, modal clustering often resorts to graph theory
for the operational detection of clusters, which is another condition that seems
particularly appropriate to deal with relational data.
