RDH: Reconstruction of Diffusion History in Cyber and Human Networks

Diffusion processes in networks can be used to model and study many real-world phenomena, including misinformation on online social networks, pandemics such as COVID-19 in human networks, and the spreading of computer viruses on the Internet. Reconstruction of diffusion history (RDH) is to identify a diffusion process that provides the best explanation of the observations where the diffusion history is a time-sequenced spreading graph. The objectives of this project include (i) establishing a new theoretical foundation for RDH based on the graph theory, stochastic processes, and statistical learning (Thrust 1), (ii) developing a new algorithmic foundation based on deep learning for RDH with partial observations (Thrust 2), and (iii) evaluating the theories and algorithms using both synthetic and real-world datasets.

Contact: Hanghang Tong