Modeling Cascading Risk in Interdependent Networks
Networked applications often operate under uncertainty in environmental response and the temporal state and action choices of the nodes are captured in the form of structured and unstructured text data as well as image data. We propose a network-centric approach that will contribute to advances in reasoning about uncertainty, large-scale text and image data analysis as well understanding of complex networks. This work is expected to lead to innovative extensions in the following research areas: combining topic modeling and information extraction from text data and image extraction for data foraging, identification of topological features for network analysis and studying the interactions between stakeholders at varying levels of the network. Specifically, we plan to study this problem in the context of Major Defense Acquisition Program (MDAP) network.
Visit the project website for the previously funded phase here.