Represent a network as a 3D shape
Mandatory arguments are mandatory for short options.
-n, network name
-f, input edge list file name.
-s, sampling proportion step (percentage)
-t, number of samples for each sampling proportion
example: python kron_hull.py -n as20graph -f as20graph.txt -s 20 -t 5
output:
the directory <network name>/
Kronecker hull <network name>/kronecker_hull.jpg
boundary points <network name>/boundary.txt
Kronecker points <network name>/kron_points.txt
edgelist files of samples
environment & prerequisite:
linux
python 2.7, networkx, numpy, scipy, matplotlib, tqdm, joblib
kronfit: please refer to https://github.com/shengminjin/snap/tree/master/examples/kronfit
To cite this work please use:
@inproceedings{jin2018representing,
title={Representing Networks with 3D Shapes},
author={Jin, Shengmin and Zafarani, Reza},
booktitle={2018 IEEE International Conference on Data Mining (ICDM)},
pages={177--186},
year={2018},
organization={IEEE}
}