Overlapping community detection on complex networks with Graph Convolutional Networks

Yuan, S.W.; Zeng, Hefeng; Zuo, Ziyang; Wang, Chao

doi:10.1016/j.comcom.2022.12.008

Cited by 20 publications

(6 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some authors [62] have employed graph based neural approaches for community detection but they were capable of detecting overlapping communities only. While some have [63,64] used concept of graph attention neural network that was capable of detecting only overlapping communities.…”

Section: A Graph Based Approachesmentioning

confidence: 99%

A Probabilistic Approach for Extractive Summarization Based on Clustering Cum Graph Ranking Method

Ahmad,

Masood

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Online information has increased tremendously in today's age of the Internet. As a result, the need has arisen to extract relevant content from the plethora of available information. Researchers are widely using automatic text summarization techniques for extracting useful and relevant information from voluminous available information. The summary obtained from the automatic text summarization often faces the issues of diversity and information coverage. Earlier researchers have used graph-based approaches for ranking and optimization. This research work introduces a probabilistic approach named as ClusRank for summary extraction, comprising of a two-stage sentence selection model involving clustering and then ranking of sentences. The initial stage involves clustering of sentences using a proposed overlapping clustering algorithm on the weighted network, and later selection of salient sentences using the introduced probabilistic approach. In the analysis of real-world networks, community structure development is essential because it provides strategic insights that help decision-makers make well-informed choices. Furthermore, methodologically strict community detection algorithms are required due to the occurrence of discontinuous, overlapping, and nested community patterns in such networks.This research work, an algorithm is presented for detecting overlapping communities based on the concept of rough set and granular information on links. The sentence selection algorithm based on budget maximum coverage approach supports the assumption that larger sub-topics in a document are of more importance than smaller subtopics. The performance of the proposed probabilistic ClusRank is validated on DUC2001, DUC 2002, DUC2004, and DUC 2006 data sets.

show abstract

Section: A Graph Based Approachesmentioning

confidence: 99%

A Probabilistic Approach for Extractive Summarization Based on Clustering Cum Graph Ranking Method

Ahmad,

Masood

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…LGCN has been applied in many recent recommendation systems with competitive results as [10], [16], and [7].…”

Section: Light Graph Convolutional Networkmentioning

confidence: 99%

ALGNet: Attention Light Graph Memory Network for Medical Recommendation System

Nguyen,

Trinh

et al. 2023

Proceedings of the 12th International Symposium on Information and Communication Technology

View full text Add to dashboard Cite

Medication recommendation is a vital task for improving patient care and reducing adverse events. However, existing methods often fail to capture the complex and dynamic relationships among patient medical records, drug efficacy and safety, and drug-drug interactions (DDI). In this paper, we propose ALGNet, a novel model that leverages light graph convolutional networks (LGCN) and augmentation memory networks (AMN) to enhance medication recommendation.LGCN can efficiently encode the patient records and the DDI graph into low-dimensional embeddings, while AMN can augment the patient representation with external knowledge from a memory module. We evaluate our model on the MIMIC-III dataset and show that it outperforms several baselines in terms of recommendation accuracy and DDI avoidance. We also conduct an ablation study to analyze the effects of different components of our model. Our results demonstrate that ALGNet can achieve superior performance with less computation and more interpretability. The implementation of this paper can be found at: https://github.com/tivivui95/ALGNet

show abstract

“…Various community detection methods have been proposed from different perspectives [4,[10][11][12], using various techniques such as autoencoder models [13][14][15], nonnegative matrix factorization [16], point-wise mutual information [17]. Learning low-dimensional representations enhances member similarity and preserves the network structure.…”

mentioning

confidence: 99%

“…Discovering the overlapping community structure within these networks holds greater practical significance as it mirrors the multifaceted relationships individuals maintain across different domains [11,25]. Considering the vast scale and intricate structures of networks, which often encompass thousands to millions of user nodes, the exploration of overlapping community detection has become a prevailing research focus.…”

mentioning

confidence: 99%

“…Moreover, these methods typically rely on prior information about the underlying community network, such as the number of communities, posing a constraint on their applicability. For instance, in [11], a community detection method based on Markov and GCN, requires prior knowledge of the number of clusters and exhibits sensitivity to graph structure. This introduces challenges in generalization, computational complexity for large graphs, assumptions related to Markov stability, and the need for the effective selection of the optimal Markov time.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Robust Overlapping Community Detection in Complex Networks With Graph Convolutional Networks and Fuzzy C-Means

Al-Andoli,

Irianto,

Alsayaydeh

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Community detection is an important task in complex network analysis. A community is a set of cohesive vertices that have more connections within the set than outside. In many real Complex Networks (CNs), these communities naturally overlap, meaning an individual node can belong to more than one community. This overlapping structure is crucial for many real applications, such as social influence detection, cyberattack detection, and recommendation systems. Existing methods often struggle to capture both network topology and node features, leading to suboptimal overlapping community detection. In this paper, we propose an efficient method called GCNFCM, which utilizes Graph Convolutional Networks (GCNs), Fuzzy C-means (FCM), and the modularity Q algorithm for overlapping community detection. The key idea is to achieve robust feature learning for nodes and then identify the best structure for overlapping community detection. GCNFCM extracts node embeddings from CNs, considering both topology and attributes through a dual-decoder design (inner product and GCN), while FCM is employed for optimal overlapping community detection. Furthermore, FCM is guided by the modularity Q algorithm for accurate community identification without requiring prior knowledge of the community count. Experimental results on ten real-world CNs of varying sizes demonstrate that our proposed method outperforms other state-of-theart overlapping community detection methods in terms of producing cohesive communities and identifying ground-truth communities. Additionally, the results indicate that the developed method effectively identifies good overlapping communities in real-world networks. INDEX TERMSGraph Convolutional Networks, Fuzzy C-means, Complex networks. Overlapping Structure, Community detection. I. INTRODUCTIONNowadays, with the rapid expansion of information technology, things in the real world are more connected than ever because and usually represented as complex networks [1]. Complex Networks (CN) are distinguished by intricate interconnections, which serve as a powerful analytical tool for understanding a variety of interconnected systems in the real world. These systems encompass diverse domains such as the internet, biological neural networks, transportation systems, and social networks [1,2]. Complex networks like social media, protein interactions, and city routes are difficult to understand because they have complicated structures and keep changing. In real-world networks, connected groups are called "communities", which share common things. Identifying these communities is crucial for understanding network structures, and is valuable in different areas like biology, sociology, and computer science. For example, in computer science, finding

show abstract

Overlapping community detection on complex networks with Graph Convolutional Networks

Cited by 20 publications

References 34 publications

A Probabilistic Approach for Extractive Summarization Based on Clustering Cum Graph Ranking Method

A Probabilistic Approach for Extractive Summarization Based on Clustering Cum Graph Ranking Method

ALGNet: Attention Light Graph Memory Network for Medical Recommendation System

Robust Overlapping Community Detection in Complex Networks With Graph Convolutional Networks and Fuzzy C-Means

Contact Info

Product

Resources

About