Networks, mathematically formalized as graphs, have become a ubiquitous abstraction of complex systems. To extract knowledge from data on such systems, the network science community has built a suite of algorithms, software packages, and visualization techniques. While the challenges posed by increasingly large datasets drive much of the research in data mining, graph mining and social network analysis, in this satellite we focus on the challenges created by increasingly complex data.

Being the continuation of successful editions held in Berkeley, CA, USA, Zaragoza, Spain, Seoul, South Korea, and Indanapolis, IN, USA this satellite will be held at NetSci 2018 in Paris. This year's edition focuses on two current research themes:

  1. Models for time series data with non-Markovian characteristics. Examples include time-resolved and sequential data on networked systems, such as vehicle trajectories, user click streams in the Web, or temporal networks. To faithfully model such data, researchers have investigated modelling techniques that are based on higher- and variable-order Markov models, which can preserve some of the rich temporal dependencies in such data.
  2. Models for complex systems that exhibit non-dyadic interactions. The key question here is to what extent the function of complex systems can be understood based on models that focus solely on pairwise interactions? This line of research focuses on higher-order modelling techniques that account for non-dyadic interactions such as simplicial complexes, hypergraphs, etc.