Prediction, modeling and control of processes in (complex) networks

Involved faculty members:

H. Wang Elvin Isufi Alan Hanjalic    

Data collected from generic complex systems such as features, relations and interactions among system components can be represented as networks. We analyze such network data and develop methodologies to address the following questions:

  • Anomaly detection, e.g. detection of the emergence of echo chamber effect in recommender systems, anomaly or risk in telecommunications network and economic networks.
  • Network analysis and embedding based link prediction. Link prediction problems range from user demand prediction in e.g. recommender systems, to the prediction of social relationships, traffic and user activities in general.
  • Process modeling and control. We model processes such as information/failure/epidemic diffusion, opinion formation, contagion of behavior that unfold upon a network and explore how network topology affects a process. We develop strategies like network modification and process intervention (via e.g. recommendations, incentivizing the activity, opinion of selected individuals) in order to e.g. reduce the prevalence of an epidemic/misinformation and dissolve the formation of social segregation and echo chambers. Our new direction AI-networking aims at automatic optimization/control solution development.

Our expertise on interdependent and time-evolving networks, thus data of interdependent and dynamic systems has been recognized by impactful publication venues, the network science, complex systems and data science communities.

Representative publications

  1. H. Wang, Q. Li, G. D'Agostino, S. Havlin, H. Eugene Stanley and P. Van Mieghem, Effect of the Interconnected Network Structure on the Epidemic Threshold, Physical Review E 88, 022801, 2013.
  2. H. Wang, C. Chen, B. Qu, D. Li, S. Havlin, Epidemic mitigation via awareness propagation in communication network: the role of time scales", New Journal of Physics 19 (7), 073039, 2017.
  3. E. Isufi, A. S. U. Mahabir and G. Leus, Blind Graph Topology Change Detection, IEEE Signal Processing Letters, vol. 15 (5), pp. 655 - 659, 2018
  4. L. Liu, B. Qu, B. Chen, A. Hanjalic and H. Wang, Modelling of information diffusion on social networks with applications to WeChat, Physica A: Statistical Mechanics and its Applications 496, 318-329, 2018.
  5. X.X. Zhan, A. Hanjalic and H. Wang, Information diffusion backbones in temporal networks, Scientific reports 9 (1), 6798, 2019.
  6. F. Gama, E. Isufi, A. Ribeiro and G. Leus, Controllability of Bandlimited Graph Processes Over Random Time-Varying Graphs, IEEE Transactions on Signal Processing, 2019.
  7. E. Altman, K. Avrachenkov, F. De Pellegrini, R. El-Azouzi and H. Wang, Multilevel Strategic Interaction Game Models for Complex Networks, Springer, 2019.
  8. X.X. Zhan, Z. Li, N. Masuda, P. Holme and H. Wang, Susceptible-infected-spreading-based network embedding in static and temporal networks, EPJ Data Science 9 (1), 30, 2020.