Deep neighbor-aware embedding for node clustering in attributed graphs

Wang, Chun; Pan, Shirui; Yu, Celina P.; Hu, Ruiqi; Long, Guodong; Zhang, Chengqi

doi:10.1016/j.patcog.2021.108230

Cited by 43 publications

(11 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Advanced embedding algorithms, e.g., for temporal networks, have been developed recently (Zhan et al 2020;Torricelli et al 2020;Tandon et al 2021). Our method can also be further applied to explore the influence of network properties on other network-based algorithms (Wang et al 2022), or dynamic processes deployed on networks (Pastor-Satorras et al 2015). More in-depth investigation of the mechanisms of embedding algorithms in relation to the network properties they could retain may further inspire the design of embedding algorithms.…”

Section: Discussionmentioning

confidence: 99%

Influence of clustering coefficient on network embedding in link prediction

et al. 2022

View full text Add to dashboard Cite

Multiple network embedding algorithms have been proposed to perform the prediction of missing or future links in complex networks. However, we lack the understanding of how network topology affects their performance, or which algorithms are more likely to perform better given the topological properties of the network. In this paper, we investigate how the clustering coefficient of a network, i.e., the probability that the neighbours of a node are also connected, affects network embedding algorithms’ performance in link prediction, in terms of the AUC (area under the ROC curve). We evaluate classic embedding algorithms, i.e., Matrix Factorisation, Laplacian Eigenmaps and node2vec, in both synthetic networks and (rewired) real-world networks with variable clustering coefficient. Specifically, a rewiring algorithm is applied to each real-world network to change the clustering coefficient while keeping key network properties. We find that a higher clustering coefficient tends to lead to a higher AUC in link prediction, except for Matrix Factorisation, which is not sensitive to the change of clustering coefficient. To understand such influence of the clustering coefficient, we (1) explore the relation between the link rating (probability that a node pair is the missing link) derived from the aforementioned algorithms and the number of common neighbours of the node pair, and (2) evaluate these embedding algorithms’ ability to reconstruct the original training (sub)network. All the network embedding algorithms that we tested tend to assign higher likelihood of connection to node pairs that share an intermediate or high number of common neighbours, independently of the clustering coefficient of the training network. Then, the predicted networks will have more triangles, thus a higher clustering coefficient. As the clustering coefficient increases, all the algorithms but Matrix Factorisation could also better reconstruct the training network. These two observations may partially explain why increasing the clustering coefficient improves the prediction performance.

show abstract

Section: Discussionmentioning

confidence: 99%

Influence of clustering coefficient on network embedding in link prediction

et al. 2022

View full text Add to dashboard Cite

show abstract

“…In another study, GMM-VGAE [19] introduces Gaussian mixture models to the variational graph autoencoder (VGAE) to capture the inherent complex data distributions, leading to the development of a unified end-to-end learning model for graph clustering. Additionally, DNENC [20] proposes a neighbor-aware GAE to gather information from neighbors, and it employs an end-to-end learning strategy. Recently, several contrastive clustering methods have also been proposed.…”

Section: Related Work 21 Graph Clusteringmentioning

confidence: 99%

“…VGAE with Gaussian mixture models (GMM-VGAE) [19] combines an autoencoder with a semi-supervised module. DNENC-Att (with a graph attentional autoencoder) and DNENC-Con (with a graph convolutional autoencoder) [20] are the most recent autoencoder methods. FGC [35] and CGC [12] are the most recent shallow methods and apply high-order structure and contrastive learning ideas, respectively, as regularizers.…”

Section: Baselinesmentioning

confidence: 99%

Preserving Global Information for Graph Clustering with Masked Autoencoders

Chen

2024

Mathematics

View full text Add to dashboard Cite

Graph clustering aims to divide nodes into different clusters without labels and has attracted great attention due to the success of graph neural networks (GNNs). Traditional GNN-based clustering methods are based on the homophilic assumption, i.e., connected nodes belong to the same clusters. However, this assumption is not always true, as heterophilic graphs are also ubiquitous in the real world, which limits the application of GNNs. Furthermore, these methods overlook global positions, which can result in erroneous clustering. To solve the aforementioned problems, we propose a novel model called Preserving Global Information for Graph Clustering with Masked Autoencoders (GCMA). We first propose a low–high-pass filter to capture meaningful low- and high-frequency information. Then, we propose a graph diffusion method to obtain the global position. Specifically, a parameterized Laplacian matrix is proposed to better control the global direction. To further enhance the learning ability of the autoencoders, we design a model with a masking strategy that enhances the learning ability. Extensive experiments on both homophilic and heterophilic graphs demonstrate GCMA’s advantages over state-of-the-art baselines.

show abstract

“…In this approach, 𝐹(𝑛) encodes both node properties and graph structure information of node n in such a way that nodes in the same cluster have similar feature vectors. This methodology, like structural methods, is separated into four primary groups based on the deep learning strategy chosen to minimize the dimension: auto Encoder-based [146][147][148][149], Generative Adversarial Network-based [112,150,151], Graph Convolutional Network-based [150,[152][153][154], and Graph Attention Network-based [119,155].…”

Section: ) Dimensional Reductionmentioning

confidence: 99%

Community Detection Algorithms in Healthcare Applications: A Systematic Review

et al. 2023

View full text Add to dashboard Cite

Over the past few years, the number and volume of data sources in healthcare databases has grown exponentially. Analyzing these voluminous medical data is both opportunity and challenge for knowledge discovery in health informatics. In the last decade, social network analysis techniques and community detection algorithms are being used more and more in scientific fields, including healthcare and medicine. While community detection algorithms have been widely used for social network analysis, a comprehensive review of its applications for healthcare in a way to benefit both health practitioners and the health informatics community is still overwhelmingly missing. This paper contributes to fill in this gap and provide a comprehensive and up-to-date literature research. Especially, categorizations of existing community detection algorithms are presented and discussed. Moreover, most applications of social network analysis and community detection algorithms in healthcare are reviewed and categorized. Finally, publicly available healthcare datasets, key challenges, and knowledge gaps in the field are studied and reviewed.

show abstract

Deep neighbor-aware embedding for node clustering in attributed graphs

Cited by 43 publications

References 8 publications

Influence of clustering coefficient on network embedding in link prediction

Influence of clustering coefficient on network embedding in link prediction

Preserving Global Information for Graph Clustering with Masked Autoencoders

Community Detection Algorithms in Healthcare Applications: A Systematic Review

Contact Info

Product

Resources

About