Workplan of Vsevolod Salnikov (Namur, Belgium)

In the last few years, the continuous development of ICT has led to a change of paradigm, sometimes called Big Data,
associated to a drastic increase in data collection methods and com- puting power. In the social sciences, it is now possible,
for the first time, to track and analyse large-scale, time-resolved data in a variety of contexts, including communication and
social net- works, and mobility. The temporal nature of the data leads to methodological challenges in order to uncover
significant temporal information and, more generally, to understand the effect of tem- porality on networked systems. Important areas of application include epidemiology, knowledge diffusion, rumour spreading, etc.
The main purpose of my STSM at ISI Torino will be to work, for a period of two weeks, on two aspects related to temporal
networks. In each case, ISI has a well-recognised expertise. On the one hand, I will focus on the problem of community
detection. This field of research has attracted much attention in recent years. If efficient methods exist for overlapping or nonoverlapping communities in static networks, the problem of finding communities in temporal networks is still a challenge.
Basic approaches consider the temporal system as a sequence of static networks where standard methods can be used. I am
currently developing a framework where I take into ac- count correlations in the time series between edges, in order to uncover overlapping, synchronised communities in networks. At ISI, an approach based on tensor factorisation has recently been developed, and the comparison between both approaches is expected to be fruitful. On the other hand, my research focuses on human mobility, an area in which ISI is active with its Socio-Patterns project. An interaction with their researchers is expected to be helpful for my sensing projects, based on different methodologies, and focusing on different resolutions. Moreover, bringing together these datasets can lead to the development of a multi scale data based proximity temporal network model.
This project continues the strand of modeling temporal networks, as presented at the First Annual Knowescape conference in
Aalto. Specifically, it contributes to work in WG 2 and 3, and could lead to methodological tools useful to track and visualise the diffusion of information, and structural re-organisation in science seen as a temporal network of collaboration and citations.

Host: Ciro Cattuto, ISI Foundation Torino