Home

Awesome

Scalable Attributed-Graph Subspace Clustering (SASGC)

This repository provides Python code to reproduce experiments from the AAAI 2023 paper Scalable Attributed-Graph Subspace Clustering.

Run Experiments

Parameter List for run.py

ParameterTypeDefaultDescription
datasetstringacmName of the graph dataset (acm, dblp, arxiv, pubmed or wiki).
powerinteger2First power to test.
runsinteger5Number of runs.

Best Propagation Orders

DatasetPropagation order
acm2
dblp2
arxiv54
computers67
wiki4
pubmed100

Example

To run the model on computers for power p=67 and have the average execution time

python run.py --dataset=computers --power 67

Citation

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}
}