Multilayer networks arise when there exists more than one source of relationship for a common set or different sets of nodes. For instance, in social networks, one can consider several types of different actors' relationships (e.g., friendship, kinship, membership, etc.). A multiplex network is a special case of multilayer network that consists in a fixed set of nodes that interacts through different types of connections. To cope with the complexity of multiplex networks, it might be useful to use tools stemming from multivariate statistics. Canonical correlation analysis and cluster analysis, for example, have been proposed to identify dimensions along which two networks are related to each other and to deal with dimension reduction in presence of several layers. In this framework, the present contribution aims at extending the use of factorial methods to visually explore the hidden structure of multiplex networks preserving the inherent complexity. More specifically, we focus on one-mode networks analyzing the corresponding set of adjacency matrices through the DISTATIS technique. This represents an extension of the multidimensional scaling applied to a set of distance matrices derived on the same set of objects. Our approach allows to represent the different kinds of relationships (inter-structures) in separate spaces and in a common space, called compromise. More specifically, it enhances the visual exploration of: i) network structure in terms of nodes' similarity in each single layer, ii) common structure of all layers, iii) nodes' variation across layers, and iv) similarity among layers' structure. How the proposed analytic procedure works will be demonstrated through a real-world data.
|Titolo:||On the Use of Multidimensional Statistical Techniques for the Analysis of Multiplex Network Data|
|Data di pubblicazione:||2017|
|Appare nelle tipologie:||4.2 Abstract|