- Rachel Domagalski, MSU
- Validating and Extending Methods for Extracting the Backbone of Bipartite Projections
- 04/29/2019
- 3:00 PM - 4:00 PM
- C517 Wells Hall
Networks are widely recognized as important for phenomena that span multiple disciplines
including the diffusion of information (communication), the contagion of illness (public health), the
passage of legislation (political science), the emergence of norms (sociology), the coordination of economic activities (economics & geography), the formation of relationships (developmental psychology), and detection of national security threats (intelligence). However, the challenges of collecting network data have limited the feasibility of network research in these areas. In this project, we will explore and refine the use of bipartite projections to validly infer otherwise unobserved or unobservable networks. Several formal methods for making such inferences have been proposed, however it remains unknown which models lead to valid inferences and under what circumstances. We use bipartite network data to systematically test, compare, and extend available models for inferring networks from bipartite projections.