A Comprehensive Survey on Community Detection with Deep Learning

@article{Su2022ACS,
  title={A Comprehensive Survey on Community Detection with Deep Learning},
  author={Xing Su and Shan Xue and Fanzhen Liu and Jia Wu and Jian Yang and Chuan Zhou and Wenbin Hu and Cecile Paris and Surya Nepal and Di Jin and Quan Z. Sheng and Philip S. Yu},
  journal={IEEE transactions on neural networks and learning systems},
  year={2022},
  volume={PP}
}
  • Xing Su, Shan Xue, Philip S. Yu
  • Published 26 May 2021
  • Computer Science
  • IEEE transactions on neural networks and learning systems
Detecting a community in a network is a matter of discerning the distinct features and connections of a group of members that are different from those in other communities. The ability to do this is of great significance in network analysis. However, beyond the classic spectral clustering and statistical inference methods, there have been significant developments with deep learning techniques for community detection in recent years--particularly when it comes to handling high-dimensional… 
A Comprehensive Survey on Deep Clustering: Taxonomy, Challenges, and Future Directions
TLDR
A comprehensive survey on deep clustering is conducted by proposing a new taxonomy of different state-of-theart approaches and summarize the essential components of deep clustered and categorize existing methods by the ways they design interactions between deep representation learning and clustering.
Region-enhanced Deep Graph Convolutional Networks for Rumor Detection
TLDR
A novel region-enhanced deep graph convolutional network (RDGCN) that enhances the propagation features of rumors by learning regionalized propagation patterns and trains to learn the propagation patterns by unsupervised learning is proposed.
A Comprehensive Survey on Graph Anomaly Detection with Deep Learning
TLDR
A taxonomy that follows a task-driven strategy and categorizes existing work according to the anomalous graph objects that they can detect is provided, and 12 future research directions spanning unsolved and emerging problems introduced by graph data, anomaly detection, deep learning and real-world applications are highlighted.
Graph-level Neural Networks: Current Progress and Future Directions
TLDR
To frame this survey, a systematic taxonomy covering GLNNs upon deep neural networks, graph Neural networks, and graph pooling is proposed and the representative and state-of-the-art models in each category are focused on this survey.
Deep Embedded Clustering with Distribution Consistency Preservation for Attributed Networks
TLDR
An end-to-end deep embedded clustering model for attributed networks is proposed that achieves significantly better or competitive performance compared with the state-of-the-art methods.
SSSNET: Semi-Supervised Signed Network Clustering
TLDR
A novel probabilistic balanced normalized cut loss for training nodes in a GNN framework for semi-supervised signed network clustering, called SSSNET, which has node clustering as main focus, with an emphasis on polarization effects arising in networks.
Discover Internet of Things
  • Cao
  • Computer Science
  • 2021
TLDR
In order to improve the feature extraction capability in convolutional process and eliminate the effect of over-smoothing problem, the bilinear information aggregator and NodeNorm normalization approaches into both encoding and decoding blocks are introduced.
ComGA: Community-Aware Attributed Graph Anomaly Detection
TLDR
A novel community-aware attributed graph anomaly detection framework (ComGA), designed to improve the performance of anomaly detection, and a tailored deep graph convolutional network (tGCN) to anomaly detection on attributed graphs.
AngHNE: Representation Learning for Bipartite Heterogeneous Networks with Angular Loss
TLDR
A novel angle-based method for bipartite heterogeneous network representation, which first construct training sets by generating quintuples, and transforms the comparisons between preferred and non-preferred samples to the comparisons of angles.
Transferring Knowledge Distillation for Multilingual Social Event Detection
TLDR
This work presents a GNN that incorporates cross-lingual word embeddings for detecting events in multilingual data streams and shows the framework to be highly effective at detection in both mult bilingual data and in languages where training samples are scarce.
...
...

References

SHOWING 1-10 OF 202 REFERENCES
A Survey of Community Detection Approaches: From Statistical Modeling to Deep Learning
TLDR
A comprehensive review of the existing community detection methods is provided and a new taxonomy that divides the existing methods into two categories, namely probabilistic graphical model and deep learning is introduced.
CommunityGAN: Community Detection with Generative Adversarial Nets
TLDR
This paper proposes CommunityGAN, a novel community detection framework that jointly solves overlapping community detection and graph representation learning and achieves substantial community detection performance gains over the state-of-the-art methods.
Deep Autoencoder-like Nonnegative Matrix Factorization for Community Detection
TLDR
Inspired by the unique feature representation learning capability of deep autoencoder, a novel model, named Deep Autoencoding-like NMF (DANMF), is proposed, named DANMF, for community detection, which can achieve better performance than the state-of-the-art NMF-based community detection approaches.
Community-Centric Graph Convolutional Network for Unsupervised Community Detection
TLDR
Experimental results on real networks showed that the new GCN approach for unsupervised community detection outperformed the best existing methods, showing the effectiveness of the novel decoding mechanism for generating links and attributes together over the commonly used methods for reconstructing links alone.
Overlapping Community Detection with Graph Neural Networks
TLDR
This work proposes a graph neural network (GNN) based model for overlapping community detection that outperforms the existing baselines by a large margin in the task of community recovery and establishes that GNN is the key ingredient to the power of the proposed model.
A Comprehensive Survey on Graph Neural Networks
TLDR
This article provides a comprehensive overview of graph neural networks (GNNs) in data mining and machine learning fields and proposes a new taxonomy to divide the state-of-the-art GNNs into four categories, namely, recurrent GNNS, convolutional GNN’s, graph autoencoders, and spatial–temporal Gnns.
Learning Community Structure with Variational Autoencoder
TLDR
A deep generative model for community detection and network generation that assumes the existence of latent similarity between nodes and allows the model to find these similarities through an automatic model selection process, capable of outperforming existing state-of-the-art methods.
Deep Learning for Community Detection: Progress, Challenges and Opportunities
TLDR
This article summarizes the contributions of the various frameworks, models, and algorithms in each stream in deep learning along with the current challenges that remain unsolved and the future research opportunities yet to be explored.
...
...