Vital nodes refer to specific nodes within a network that can exert a greater influence on the structure and functionality of the network compared to other nodes. During the past decades, people mainly focus on developing new vital node identification algorithms. However, there is still no consensus on how to evaluate and compare these methods. This is mainly due to the lack of standard benchmark networks with known ground-truth vital nodes.
To fill this crucial gap, a research team led by Zengyou He published their new research on 15 October 2025 in Frontiers of Computer Science co-published by Higher Education Press and Springer Nature.
The team developed BG 2 VN, a MATLAB toolbox designed for generating benchmark graphs that are composed of ground-truth vital nodes. BG 2 VN is based on the two-dimensional Gaussian distribution. It generates an adjacency matrix containing a specified number of vital nodes by methods such as restricting the distance between samples of the two-dimensional Gaussian distribution, setting the probability of edge generation between two points, and imposing constraints on edge generation
BG 2 VN provides a standardized evaluation benchmark for vital node mining algorithms. It allows for adjustments to the graph’s heterogeneity and aggregation based on user requirements.
Frontiers of Computer Science
Experimental study
Not applicable
BG^2VN: benchmark graph generator for vital node recognition
15-Oct-2025