This project aims to develop a comprehensive mathematical framework for modeling disinformation propagation in networked environments through the integration of statistical analysis and temporal graph neural networks (TGNNs). The research will conceptualize information diffusion as a dynamic process on evolving graphs, enhancing traditional epidemic-style diffusion models with the more flexible temporal point process framework to better capture the stochastic nature of information transmission across diverse network topologies.

The candidate will investigate how temporal point processes can effectively model the sporadic and bursty nature of content sharing, enabling precise mathematical formulation of cascade effects in both centralized and decentralized social structures. Adaptations of clustering techniques will be explored to identify emergent patterns in user behavior, and novel mathematical metrics will be developed to quantify the influence potential of network nodes.

Temporal graph neural network architectures will be incorporated to learn time-varying node embeddings that capture the evolutionary dynamics of information spread. The research will establish theoretical bounds on disinformation propagation rates and develop predictive models that quantify uncertainty through rigorous statistical inference.