Modularity-aware graph autoencoders for joint community detection and link prediction

Salha-Galvan, Guillaume; Lutzeyer, Johannes F.; Dasoulas, George; Hennequin, Romain; Vazirgiannis, Michalis

doi:10.1016/j.neunet.2022.06.021

Cited by 36 publications

(11 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our work paves the way for various future research, including replacing Louvain with other prior methods, using our regularizer in conjunction with other reconstruction losses (e.g., ELBO variants computed from Gaussian mixtures [7,8]), and extending our approach to dynamic graphs. The journal version 2 of this work [44] includes several additional extensions as well as results, omitted here for brevity. This includes further comparisons to non-GAE/VGAE methods, a spectral analysis of our message passing operator, and discussions on how this research helps the music streaming service Deezer address real-world multi-task LP and CD problems for music recommendation purposes.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

New Frontiers in Graph Autoencoders: Joint Community Detection and Link Prediction

Salha-Galvan¹,

Lutzeyer²,

Dasoulas³

et al. 2022

Preprint

View full text Add to dashboard Cite

Graph autoencoders (GAE) and variational graph autoencoders (VGAE) emerged as powerful methods for link prediction (LP). Their performances are less impressive on community detection (CD), where they are often outperformed by simpler alternatives such as the Louvain method. It is still unclear to what extent one can improve CD with GAE and VGAE, especially in the absence of node features. It is moreover uncertain whether one could do so while simultaneously preserving good performances on LP in a multi-task setting. In this workshop paper, summarizing results from our journal publication [44], we show that jointly addressing these two tasks with high accuracy is possible. For this purpose, we introduce a communitypreserving message passing scheme, doping our GAE and VGAE encoders by considering both the initial graph and Louvain-based prior communities when computing embedding spaces. Inspired by modularity-based clustering, we further propose novel training and optimization strategies specifically designed for joint LP and CD. We demonstrate the empirical effectiveness of our approach, referred to as Modularity-Aware GAE and VGAE, on various real-world graphs.

show abstract

Section: Discussionmentioning

confidence: 99%

“…This workshop paper summarizes results from our journal article "Modularity-Aware Graph Autoencoders for Joint Community Detection and Link Prediction" accepted for publication in Elsevier's Neural Networks journal in 2022[44]. The purpose of our submission to GLFrontiers was to present this work to a live audience 3.…”

mentioning

confidence: 99%

New Frontiers in Graph Autoencoders: Joint Community Detection and Link Prediction

Salha-Galvan¹,

Lutzeyer²,

Dasoulas³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The concept of modularity is based on a comparison between the compactness of connections within and across communities. The following equation returns a scalar value of Q in the range [−1, 1]; hence, a higher value is recommended (Liang et al , 2016; Salha-Galvan et al , 2022). where A denotes the adjacency matrix, “m” represents a number of edges in a graph, K i denotes degree ( i ) and δ( S i ,S j ) defines the probability of nodes that belongs to the same community.…”

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

CommunityGCN: community detection using node classification with graph convolution network

Bhattacharya

Nagwani²,

Tripathi³

2023

DTA

View full text Add to dashboard Cite

PurposeA community demonstrates the unique qualities and relationships between its members that distinguish it from other communities within a network. Network analysis relies heavily on community detection. Despite the traditional spectral clustering and statistical inference methods, deep learning techniques for community detection have grown in popularity due to their ease of processing high-dimensional network data. Graph convolutional neural networks (GCNNs) have received much attention recently and have developed into a potential and ubiquitous method for directly detecting communities on graphs. Inspired by the promising results of graph convolutional networks (GCNs) in analyzing graph structure data, a novel community graph convolutional network (CommunityGCN) as a semi-supervised node classification model has been proposed and compared with recent baseline methods graph attention network (GAT), GCN-based technique for unsupervised community detection and Markov random fields combined with graph convolutional network (MRFasGCN).Design/methodology/approachThis work presents the method for identifying communities that combines the notion of node classification via message passing with the architecture of a semi-supervised graph neural network. Six benchmark datasets, namely, Cora, CiteSeer, ACM, Karate, IMDB and Facebook, have been used in the experimentation.FindingsIn the first set of experiments, the scaled normalized average matrix of all neighbor's features including the node itself was obtained, followed by obtaining the weighted average matrix of low-dimensional nodes. In the second set of experiments, the average weighted matrix was forwarded to the GCN with two layers and the activation function for predicting the node class was applied. The results demonstrate that node classification with GCN can improve the performance of identifying communities on graph datasets.Originality/valueThe experiment reveals that the CommunityGCN approach has given better results with accuracy, normalized mutual information, F1 and modularity scores of 91.26, 79.9, 92.58 and 70.5 per cent, respectively, for detecting communities in the graph network, which is much greater than the range of 55.7–87.07 per cent reported in previous literature. Thus, it has been concluded that the GCN with node classification models has improved the accuracy.

show abstract

“…Identifying missing edges from the unobserved edges in a network is the objective of the link prediction task [4]. To achieve better performance in the link prediction task, there is increasing interest among many researchers to utilize GNN models in their studies [5][6][7][8][9][10][11][12].…”

Section: Introductionmentioning

confidence: 99%

Improving Graph Neural Network Models in Link Prediction Task via A Policy-Based Training Method

et al. 2022

View full text Add to dashboard Cite

Graph neural network (GNN), as a widely used deep learning model in processing graph-structured data, has attracted numerous studies to apply it in the link prediction task. In these studies, observed edges in a network are utilized as positive samples, and unobserved edges are randomly sampled as negative ones. However, there are problems in randomly sampling unobserved edges as negative samples. First, some unobserved edges are missing edges that are existing edges in the network. Second, some unobserved edges can be easily distinguished from the observed edges, which cannot contribute sufficiently to the prediction task. Therefore, using the randomly sampled unobserved edges directly as negative samples is difficult to make GNN models achieve satisfactory prediction performance in the link prediction task. To address this issue, we propose a policy-based training method (PbTRM) to improve the quality of negative samples. In the proposed PbTRM, a negative sample selector generates the state vectors of the randomly sampled unobserved edges and determines whether to select them as negative samples. We perform experiments with three GNN models on two standard datasets. The results show that the proposed PbTRM can enhance the prediction performance of GNN models in the link prediction task.

show abstract

Modularity-aware graph autoencoders for joint community detection and link prediction

Cited by 36 publications

References 49 publications

New Frontiers in Graph Autoencoders: Joint Community Detection and Link Prediction

New Frontiers in Graph Autoencoders: Joint Community Detection and Link Prediction

CommunityGCN: community detection using node classification with graph convolution network

Improving Graph Neural Network Models in Link Prediction Task via A Policy-Based Training Method

Contact Info

Product

Resources

About