Mapping Brexit debate on twitter via functional graphical models

In recent years a literature on multivariate functional graph models has been developed. The graphical representation of the conditional dependence among a finite number of random variables is indeed appealing in different applications, such as for example the analysis of the brain connectivity. We want to investigate a novel extension of this methodology, considering random functions spatially and temporally correlated. A motivating example is the analysis of the semantic network emerging from twitter users. In particular the main goal of our analysis is to track the change of the Brexit debate on Twitter across UK during a particular time frame. By considering the change in time of a word usage as a functional realization, the semantic network regarding the topic of interest is then defined as a graphical representation of the conditional dependence among functional variables. Since each tweet considered is localized in both time and space we shall take into accounts such features to properly define the functional semantic network.