Abstract:We discuss efficient computation of $f$-centralities and nonbacktracking centralities for time-evolving networks with nonnegative weights. We present a node-level formula for its combinatorially exact computation which proves to be more tractable than previously existing formulae at edge-level for dense networks. Additionally, we investigate the impact of the addition of a final time frame to such a time-evolving network, analyzing its effect on the resulting nonbacktracking Katz centrality. Finally, we demonstrate by means of computational experiments that the node-level formula presented is much more efficient for dense networks than the previously known edge-level formula. As a tool for our goals, in an appendix of the paper, we develop a spectral theory of matrices whose elements are vectors.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: **How to efficiently calculate f - centrality and non - backtracking Katz centrality in time - varying networks**. Specifically, the authors propose a node - level formula to calculate these centrality measures, which is more efficient in dense networks than the existing edge - level methods. ### Problem Background 1. **Time - varying Characteristics of Complex Networks** - In many fields, complex networks change as they evolve over time, and this time - dependence makes the network structure more complex. - To deal with these time - varying networks, the time interval is usually discretized into multiple time stamps (or time frames), and each time stamp corresponds to a static network. 2. **Importance of Centrality Measures** - Centrality measures are crucial tools for analyzing the importance of nodes in a network. - Common centrality measures include path - combination - based measures, such as Katz centrality, which measures the importance of a node by weighting all paths starting from the node. 3. **Limitations of Existing Methods** - For time - varying networks, calculating f - centrality and non - backtracking Katz centrality usually requires generating the adjacency matrix of the line graph or other edge - related matrices, and this method has a very high computational cost in dense networks. - Especially in fields such as financial mathematics, time - varying networks are often very dense, so more efficient calculation methods are required. ### Main Contributions of the Paper 1. **Efficient Node - level Formula** - The authors propose a node - level formula to calculate f - centrality and non - backtracking Katz centrality in time - varying networks. - This method uses the form of block matrices, avoiding the generation of large - scale line - graph matrices, thus improving computational efficiency. 2. **Theoretical Derivation and Experimental Verification** - The paper proves the correctness and effectiveness of the new formula through strict mathematical derivations. - The experimental results show that for dense networks, the node - level formula is more efficient than the traditional edge - level formula. 3. **Application Prospects** - This method is particularly suitable for research in fields such as financial mathematics, for example, by constructing time - varying networks to optimize portfolio selection. ### Key Formulas 1. **Calculation Formula for f - centrality** \[ \text{f - centrality of node } i=\sum_{j = n(\tau_1 - 1)+1}^{\tau_2 n}[f(tA)]_{(\tau_1 - 1)n + i,j} \] where $ f(t)=\sum_{k = 0}^{\infty}c_k t^k $, and $ A $ is the node - level adjacency matrix of the time - varying network. 2. **Generating Function for Non - backtracking Katz Centrality** \[ \Psi(A,t)=\left[I - tA*(E - t^2 A^T*A)^{-1}+\text{dd}^*(tA(tA*(E - t^2 A^T*A)^{-1}))\right]^{-1} \] where $ A $ is the node - level adjacency matrix of the time - varying network, and $ \text{dd}^* $ represents the diagonal matrix operation. Through these improvements, the paper provides a more efficient method for calculating centrality measures in time - varying networks, especially performing excellently in dense networks.

Efficient computation of \lowercase{$f$}-centralities and nonbacktracking centrality for temporal networks

EIGENVECTOR-BASED CENTRALITY MEASURES FOR TEMPORAL NETWORKS

Temporal Walk Centrality: Ranking Nodes in Evolving Networks

Efficient computation of Katz centrality for very dense networks via negative parameter Katz

Updating Katz centrality by counting walks

Efficient Exact and Approximate Betweenness Centrality Computation for Temporal Graphs

Tunable Eigenvector-Based Centralities for Multiplex and Temporal Networks

Faster Betweenness Centrality Updates in Evolving Networks

TIME CENTRALITY IN DYNAMIC COMPLEX NETWORKS

Labeling-based centrality approaches for identifying critical edges on temporal graphs

Supracentrality Analysis of Temporal Networks with Directed Interlayer Coupling

Temporal betweenness centrality in dynamic graphs

Identifying Vital Nodes On Temporal Networks: An Edge-Based K-Shell Decomposition

Graph Metrics for Temporal Networks

The localization of non-backtracking centrality in networks and its physical consequences

Eigenvector-based centralities for multilayer temporal networks under the framework of tensor computation

Using Time-Aware Graph Neural Networks to Predict Temporal Centralities in Dynamic Graphs

Temporal Betweenness Centrality on Shortest Walks Variants

Higher-Order Aggregate Networks in the Analysis of Temporal Networks: Path structures and centralities

Temporal-varying failures of nodes in networks

On the sensitivity of centrality metrics