D2.2. Report on correlation methods

In this report we review recent advances in the analysis of fully connected networks with weighted links. The weight of a link can represent the similarity between a pair of nodes, e.g., a correlation, or their dissimilarity, e.g., a distance. While for the former several approaches have been developed to uncover the underlying topology, for the latter there isn’t a standard procedure so far. An approach consists in mapping the weights of the links to a probability in the range [0,1]. Generalized measures defined in terms of the probabilities can then be applied. An alternative approach introduces a threshold allowing converting the weighted networks into directed networks with binary‐valued adjacency matrices.

^TOP