This repository provides Python code to reproduce experiments from the AAAI 2023 paper Scalable Attributed-Graph Subspace Clustering.
Parameter | Type | Default | Description |
---|---|---|---|
dataset |
string | acm |
Name of the graph dataset (acm , dblp , arxiv , pubmed or wiki ). |
power |
integer | 2 |
First power to test. |
runs |
integer | 5 |
Number of runs. |
Dataset | Propagation order |
---|---|
acm |
2 |
dblp |
2 |
arxiv |
54 |
computers |
67 |
wiki |
4 |
pubmed |
100 |
To run the model on computers for power p=67
and have the average execution time
python run.py --dataset=computers --power 67
If you use this code please do cite :
@inproceedings{fettal2023scalable,
title={Scalable Attributed-Graph Subspace Clustering},
author={Fettal, Chakib and Labiod, Lazhar and Nadif, Mohamed},
booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
volume={37},
year={2023}
}