Identifying vital nodes on temporal networks: An edge-based K-shell decomposition

There is an ever-increasing interest in studying temporal networks nowadays, as temporal networks can illustrate the real-world system more accurately. To date, how to characterize nodes' importance is still unclear in temporal networks. In this work, we first use a time window graph model to cut the temporal network into slices, and then we give an indicator for network centrality according to the edge-based κ-shell decomposition for the temporal networks, which is named as temporal κ-shell decomposition. We mainly use the size of the largest component after removing the nodes with large centrality value to test the method's performance. The numerical experiments on several real networks indicate that the temporal κ-shell method outperforms some other indicators, and the results with different time window size show that the improvement is also robust.