Predicting the structure of dynamic graphs
This addresses a gap in graph analysis for dynamic systems, but it is incremental as it combines existing methods like time series forecasting and flux balance analysis.
The paper tackles the problem of forecasting future graph structures, including unseen nodes and edges, using a time series of graphs, and demonstrates its utility on synthetic and real-world datasets.
Many aspects of graphs have been studied in depth. However, forecasting the structure of a graph at future time steps incorporating unseen, new nodes and edges has not gained much attention. In this paper, we present such an approach. Using a time series of graphs, we forecast graphs at future time steps. We use time series forecasting methods to predict the node degree at future time points and combine these forecasts with flux balance analysis -- a linear programming method used in biochemistry -- to obtain the structure of future graphs. We evaluate this approach using synthetic and real-world datasets and demonstrate its utility and applicability.