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 yueliu19990731@163.com. If you find this repository useful to your research or work, it is really appreciated to star this repository. ✨ If you use our code or the processed datasets in this repository for your research, please cite 1-2 papers in the citation part. ❤️

---

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]
2. "A Comprehensive Survey on Community Detectionwith Deep Learning" [paper]
3. "A Comprehensive Survey on Graph Neural Networks" [paper]
4. "A Survey on Graph Structure Learning: Progress and Opportunities" [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. AdaGAE (TPAMI 21): "Adaptive Graph Auto-Encoder for General Data Clustering" [paper|code]
7. TCL (IJCV 22): "Twin Contrastive Learning for Online Clustering" [paper|code]

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. AHGAE (TKDE 21): "Adaptive Hypergraph Auto-Encoder for Relational Data Clustering" [paper]
17. EGAE (TNNLS 22): "Embedding Graph Auto-Encoder for Graph Clustering"[paper|code]
18. GC-VGE (PR 22): "Graph Clustering via Variational Graph Embedding" [paper]
19. GCC (WSDM 22): "Efficient Graph Convolution for Joint Node Representation Learning and Clustering" [paper|code]
20. scTAG (AAAI 22): "ZINB-based Graph Embedding Autoencoder for Single-cell RNA-seq Interpretations" [paper|code]
21. CDRS (TNNLS 22): "Collaborative Decision-Reinforced Self-Supervision for Attributed Graph Clustering" [paper|code]
22. DNENC (PR 22): "Deep neighbor-aware embedding for node clustering in attributed graphs" [paper]

Adversarial Deep Graph Clustering

1. ARGA/ARVGA (IJCAI 18) : "Adversarially Regularized Graph Autoencoder for Graph Embedding" [paper|code]
2. AGAE (IJCAI 19): "Adversarial Graph Embedding for Ensemble Clustering" [paper]
3. ARGA/ARVGA (TCYB 19): "Learning Graph Embedding with Adversarial Training Methods" [paper|code]
4. WARGA (arXiv 21): "Wasserstein Adversarially Regularized Graph Autoencoder" [paper|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. DCRN (AAAI 22): "Deep Graph Clustering via Dual Correlation Reduction" [paper|code]
8. AFGRL (AAAI 22): "Augmentation-Free Self-Supervised Learning on Graphs" [paper|code]
9. SAIL(AAAI 22): "SAIL: Self-Augmented Graph Contrastive Learning" [paper]
10. IDCRN (arXiv 22, extension of DCRN): "Improved Dual Correlation Reduction Network" [paper]
11. AGC-DRR (IJCAI 22): "Attributed Graph Clustering with Dual Redundancy Reduction" [paper|code]
12. SUBLIME (WWW 22): "Towards Unsupervised Deep Graph Structure Learning" [paper|code]
13. CGC (WWW 22): "CGC: Contrastive Graph Clustering for Community Detection and Tracking" [paper]
14. SCGC (arXiv 22): "Simple Contrastive Graph Clustering" [paper]
15. SCGC (arXiv 22): "SCGC : Self-Supervised Contrastive Graph Clustering" [paper|code]
16. NCAGC (arXiv 22): "NCAGC: A Neighborhood Contrast Framework for Attributed Graph Clustering" [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. SSGC (ICLR 21): "Simple Spectral Graph Convolution" [paper|code]
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]
27. GRACE (TKDD 22): "GRACE: A General Graph Convolution Framework for Atributed Graph Clustering" [paper]

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
   CITE	3327	3703	4552	6	cite.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	17	wiki.zip
   COCS	18333	6805	81894	15	cocs.zip
   CORNELL	183	1703	149	5	cornell.zip
   TEXAS	183	1703	162	5	texas.zip
   WISC	251	1703	257	5	wisc.zip
   FILM	7600	932	15009	5	film.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

@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={Proceedings of the AAAI Conference on Artificial Intelligence},
  volume={36},
  number={7},
  pages={7603-7611},
  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