ADGC: Awesome Deep Graph Clustering

ADGC is a collection of state-of-the-art (SOTA), novel deep graph clustering methods (papers, codes and datasets). Any other interesting papers and codes are welcome. Any problems, please contact [email protected]. If you find this repository useful to your research or work, it is really appreciated to star this repository. ❤️

---

What's Deep Graph Clustering?

Deep graph clustering, which aims to reveal the underlying graph structure and divide the nodes into different groups, has attracted intensive attention in recent years.

​

Important Survey Papers

1. "A survey of clustering with deep learning: From the perspective of network architecture" [paper]

Papers

Traditional Clustering

1. K-Means: "Algorithm AS 136: A k-means clustering algorithm" [paper|code]

Deep Clustering

1. DEC (ICML 16): "Unsupervised Deep Embedding for Clustering Analysis" [paper|code]
2. IDEC (IJCAI 17): "Improved Deep Embedded Clustering with Local Structure Preservation" [paper|code]
3. DCN (ICML 17): "Towards k-means-friendly spaces: Simultaneous deep learning and clustering" [paper|code]
4. CC (AAAI 21): "Contrastive Clustering" [paper|code]
5. Ada-Net (ICLR 22): "Ada-nets: Face clustering via adaptive neighbour discovery in the structure space" [paper|code]
6. SAIL(AAAI 22): "SAIL: Self-Augmented Graph Contrastive Learning" [paper]

Generative Deep Graph Clustering

1. SAE (AAAI 14): "Learning Deep Representations for Graph Clustering" [paper|code]
2. GAE/VGAE (NeurIPS 16): "Variational Graph Auto-Encoders" [paper|code]
3. MGAE (CIKM 17): "MGAE: Marginalized Graph Autoencoder for Graph Clustering" [paper|code]
4. NetVAE (IJCAI 19): "Network-Specific Variational Auto-Encoder for Embedding in Attribute Networks" [paper]
5. DAEGC (IJCAI 19): "Attributed Graph Clustering: A Deep Attentional Embedding Approach" [paper|code]
6. GALA (ICCV 19): "Symmetric Graph Convolutional Autoencoder for Unsupervised Graph Representation Learning" [paper|code]
7. RWR-GAE (arXiv 19): "RWR-GAE: Random Walk Regularization for Graph Auto Encoders" [paper|code]
8. DGVAE (NeurIPS 20): "Dirichlet Graph Variational Autoencoder" [paper|code]
9. SDCN/SDCN_Q (WWW 20): "Structural Deep Clustering Network" [paper|code]
10. O2MAC (WWW 20): "One2Multi Graph Autoencoder for Multi-view Graph Clustering" [paper|code]
11. CGCN/GMM-VGAE (AAAI 20): "Collaborative Graph Convolutional Networks: Unsupervised Learning Meets Semi-Supervised Learning" [paper|code]
12. DFCN (AAAI 21): "Deep Fusion Clustering Network" [paper|code]
13. R-GAE (arXiv 21): "Rethinking Graph Auto-Encoder Models for Attributed Graph Clustering" [paper|code]
14. AGCN (MM 21): "Attention-driven Graph Clustering Network" [paper|code]
15. DAGC (arXiv 21, extension of AGCN): ”Deep Attention-guided Graph Clustering with Dual Self-supervision“ [paper|code]
16. AdaGAE (TPAMI 21): "Adaptive Graph Auto-Encoder for General Data Clustering" [paper|code]
17. AHGAE (TKDE 21): "Adaptive Hypergraph Auto-Encoder for Relational Data Clustering" [paper]
18. EGAE (TNNLS 22): "Embedding Graph Auto-Encoder for Graph Clustering"[paper|code]
19. GC-VGE (PR 22): "Graph Clustering via Variational Graph Embedding" [paper]
20. GCC (WSDM 22): "Efficient Graph Convolution for Joint Node Representation Learning and Clustering" [paper|code]
21. scTAG (AAAI 22): "ZINB-based Graph Embedding Autoencoder for Single-cell RNA-seq Interpretations" [paper|code]

Adversarial Deep Graph Clustering

1. AGAE (IJCAI 19): "Adversarial Graph Embedding for Ensemble Clustering" [paper]
2. ARGA/ARVGA (TCYB 19): "Learning Graph Embedding with Adversarial Training Methods" [paper|code]
3. WARGA (arXiv 21): "Wasserstein Adversarially Regularized Graph Autoencoder" [paper|code]
4. AGC-DRR (IJCAI 22): "Attributed Graph Clustering with Dual Redundancy Reduction" [code]

Contrastive Deep Graph Clustering

1. MVGRL (ICML 20): "Contrastive Multi-View Representation Learning on Graphs" [paper|code]
2. AGE (SIGKDD 20): "Adaptive Graph Encoder for Attributed Graph Embedding" [paper|code]
3. MCGC (NeurIPS 21): "Multi-view Contrastive Graph Clustering" [paper|code]
4. SCAGC (arXiv 21): "Self-supervised Contrastive Attributed Graph Clustering" [paper]
5. MGCCN (arXiv 21): "Multilayer Graph Contrastive Clustering Network" [paper]
6. GDCL (IJCAI 21): "Graph Debiased Contrastive Learning with Joint Representation Clustering" [paper|code]
7. GCC (ICCV 21): "Graph Contrastive Clustering" [paper|code]
8. DCRN (AAAI 22): "Deep Graph Clustering via Dual Correlation Reduction" [paper|code]
9. AFGRL (AAAI 22): "Augmentation-Free Self-Supervised Learning on Graphs" [paper|code]
10. IDCRN (arXiv 22, extension of DCRN): "Improved Dual Correlation Reduction Network" [paper]
11. SUBLIME (WWW 22): "Towards Unsupervised Deep Graph Structure Learning" [paper|code]

Other Related Methods

1. AGC (IJCAI 19): "Attributed Graph Clustering via Adaptive Graph Convolution" [paper|code]
2. MCGC (TIP 19): "Multiview Consensus Graph Clustering" [paper|code]
3. HAN (WWW 19): "Heterogeneous Graph Attention Network" [paper|code]
4. DAG (ICCKE 20): "Attributed Graph Clustering via Deep Adaptive Graph Maximization" [paper]
5. MAGCN (IJCAI 20): "Multi-View Attribute Graph Convolution Networks for Clustering" [paper|code]
6. Cross-Graph (ICDM 20): "Cross-Graph: Robust and Unsupervised Embedding for Attributed Graphs with Corrupted Structure" [paper|code]
7. DMGI (AAAI 20): "Unsupervised Attributed Multiplex Network Embedding" [paper|code]
8. MAGNN (WWW 20): "MAGNN: Metapath Aggregated Graph Neural Network for Heterogeneous Graph Embedding" [paper|code]
9. MinCutPool (ICML 20): "Spectral Clustering with Graph Neural Networks for Graph Pooling" [paper|code]
10. GCAGC (CVPR 20): "Adaptive Graph Convolutional Network with Attention Graph Clustering for Co saliency Detection" [paper|code]
11. DBGAN (CVPR 20): "Distribution-induced Bidirectional GAN for Graph Representation Learning" [paper|code]
12. CAGNN (arXiv 20): "Cluster-Aware Graph Neural Networks for Unsupervised Graph Representation Learning" [paper]
13. COLES (NeurIPS 21): "Contrastive Laplacian Eigenmaps" [paper|code]
14. Graph-MVP (arXiv 21): "Graph-MVP: Multi-View Prototypical Contrastive Learning for Multiplex Graphs" [paper|code]
15. MvAGC (IJCAI 21): "Graph Filter-based Multi-view Attributed Graph Clustering" [paper|code]
16. GIC (PAKDD 21): "Graph InfoClust: Maximizing Coarse-Grain Mutual Information in Graphs" [paper|code]
17. HDMI (WWW 21): "High-order Deep Multiplex Infomax" [paper|code]
18. MAGC (TKDE 21): "Multi-view Attributed Graph Clustering" [paper|code]
19. NAS-GC (TCYB 21): "Smoothness Sensor: Adaptive Smoothness Transition Graph Convolutions for Attributed Graph Clustering" [paper|code]
20. SENet (NN 21): "Spectral embedding network for attributed graph clustering" [paper]
21. DNENC (PR 22): "Deep neighbor-aware embedding for node clustering in attributed graphs" [paper]
22. SDSNE(AAAI 22): "Stationary diffusion state neural estimation for multiview clustering" [paper|code]
23. AUTOSSL (ICLR 22): "Automated Self-Supervised Learning For Graphs" [paper|code]
24. SAGES (TKDE 22): "SAGES: Scalable Attributed Graph Embedding with Sampling for Unsupervised Learning" [paper]
25. MVGC (NN 22): "Multi-view graph embedding clustering network: Joint self-supervision and block diagonal representation" [paper|code]
26. FGC (SDM22): "Fine-grained Attributed Graph Clustering" [paper|code]

Benchmark Datasets

We divide the datasets into two categories, i.e. graph datasets and non-graph datasets. Graph datasets are some graphs in real-world, such as citation networks, social networks and so on. Non-graph datasets are NOT graph type. However, if necessary, we could construct "adjacency matrices" by K-Nearest Neighbors (KNN) algorithm.

Quick Start

• Step1: Download all datasets from [Google Drive | Nutstore]. Optionally, download some of them from URLs in the tables (Google Drive)
• Step2: Unzip them to ./dataset/
• Step3: Change the type and the name of the dataset in main.py
• Step4: Run the main.py

Code

• utils.py
  1. load_graph_data: load graph datasets
  2. load_data: load non-graph datasets
  3. normalize_adj: normalize the adjacency matrix
  4. diffusion_adj: calculate the graph diffusion
  5. construct_graph: construct the knn graph for non-graph datasets
  6. numpy_to_torch: convert numpy to torch
  7. torch_to_numpy: convert torch to numpy
• clustering.py
  1. setup_seed: fix the random seed
  2. evaluation: evaluate the performance of clustering
  3. k_means: K-means algorithm
• visualization.py
  1. t_sne: t-SNE algorithm
  2. similarity_plot: visualize cosine similarity matrix of the embedding or feature

Datasets Details

About the introduction of each dataset, please check here

1. Graph Datasets

   Dataset	Samples	Dimension	Edges	Classes	URL
   CORA	2708	1433	5278	7	cora.zip
   CITESEER	3327	3703	4552	6	citeseer.zip
   PUBMED	19717	500	44324	3	pubmed.zip
   DBLP	4057	334	3528	4	dblp.zip
   ACM	3025	1870	13128	3	acm.zip
   AMAP	7650	745	119081	8	amap.zip
   AMAC	13752	767	245861	10	amac.zip
   CORAFULL	19793	8710	63421	70	corafull.zip
   WIKI	2405	4973	8261	19	wiki.zip
   COCS	18333	6805	81894	15	cocs.zip
   BAT	131	81	1038	4	bat.zip
   EAT	399	203	5994	4	eat.zip
   UAT	1190	239	13599	4	uat.zip

Edges: Here, we just count the number of undirected edges.

2. Non-graph Datasets

   Dataset	Samples	Dimension	Type	Classes	URL
   USPS	9298	256	Image	10	usps.zip
   HHAR	10299	561	Record	6	hhar.zip
   REUT	10000	2000	Text	4	reut.zip

Citation

If you use our code or the processed datasets in this repository for your research, please cite our paper.

@inproceedings{DCRN,
  title={Deep Graph Clustering via Dual Correlation Reduction},
  author={Liu, Yue and Tu, Wenxuan and Zhou, Sihang and Liu, Xinwang and Song, Linxuan and Yang, Xihong and Zhu, En},
  booktitle={AAAI Conference on Artificial Intelligence},
  year={2022}
}
@article{mrabah2021rethinking,
  title={Rethinking Graph Auto-Encoder Models for Attributed Graph Clustering},
  author={Mrabah, Nairouz and Bouguessa, Mohamed and Touati, Mohamed Fawzi and Ksantini, Riadh},
  journal={arXiv preprint arXiv:2107.08562},
  year={2021}
}

Other Related Awesome Repository

awesome-awesome-machine-learning

Awesome Incomplete Graph Representation Learning

awesome-self-supervised-gnn