Identifying key players in complex networks through network entanglement

Yiming Huang, Hao Wang, Xiao-Long Ren, Linyuan Lü. Identifying key players in complex networks through network entanglement. Communications Physics (2024).

This paper proposed an entanglement-based metric - vertex entanglement (VE) - quantifying local perturbations on spectral entropy, with superior applications in network dismantling and brain network analysis.

Repo Contents

1. data
   • please put your dataset in the following directory './data/{data_name}/edges.txt'
2. src: source code of VE
   • VE.py: python version implementation of our algorithm.
   • reinsertion.py: this algorithm reinserts nodes for network dismantling.
   • GNDR.py: The original GNDR code is not in Python, and it is not adapted for non-connected graphs. We provide a Python version of the GNDR algorithm.
3. utils: basic utils used in the src code

Reproduce results

Run VE.py directly to reproduce the dismantling results in our paper.

• The resulting list of removing nodes is given in './results/{data_name}/VER_nodelist.txt'
• The resulting GCC change list corresponding to the given removing nodes list is reported in './results/{data_name}/VER_gcc.txt'
• Change netname in VE.py to conduct experiments for different datasets. Note that your dataset should put the edge list in the following directory './data/{data_name}/edges.txt'

Basebline methods implementation

We used the source codes released online, and adopted the best parameter settings provided by authors (if available) for each method.

https://github.com/zhfkt/ComplexCi (CI)
https://github.com/abraunst/decycler (MinSum)
http://power.itp.ac.cn/~zhouhj/codes.html (BPD)
https://github.com/hcmidt/corehd (CoreHD)
https://github.com/renxiaolong/Generalized-Network-Dismantling (GNDR)

Reference

Please cite our work if you find our code/paper is useful to your work:

@article{VE2024Huang,  
  title = {Identifying key players in complex networks via network entanglement},  
  author={Huang, Yiming and Wang, Hao and Ren, Xiao-Long and L{\"u}, Linyuan},
  journal = {Communications Physics},  
  year = {2024},  
  volume = {7},  
  number = {1},  
  pages = {19},  
  issn = {2399-3650},  
  url = {https://doi.org/10.1038/s42005-023-01483-8},  
  doi = {10.1038/s42005-023-01483-8},  
}