Multiplex Network Reconstruction

This is the implementation of numerical experiments in our paper titled Reconstructing Sparse Multiplex Networks with Application to Covert Networks. If you find our paper or code useful, we kindly ask you to cite our work

@article{yu2023reconstructing,
  title={Reconstructing sparse multiplex networks with application to covert networks},
  author={Yu, Jin-Zhu and Wu, Mincheng and Bichler, Gisela and Aros-Vera, Felipe and Gao, Jianxi},
  journal={Entropy},
  volume={25},
  number={1},
  pages={142},
  year={2023},
  publisher={MDPI}
}

Requirements

Software

Python 3.8.3

Packages

numpy 1.19.2

pandas 1.4.4

numba 0.55.1

sklearn 1.1.1

imblearn 0.7.0

networkx 2.8.4

Usage

To reproduce the plots for reconstructuring each multiplex network, run multi_net.py and then plot_metrics.py with the respective net_name, n_layer, and n_node_total.

Parellel processing is used to reduce the runtime. If necessary, Cython can be used to decrease the runtime a bit more.