STEC-Net: A Spatiotemporal Graph Neural Framework for Community Discovery in Dynamic Social Networks

Shuangshuang Chu; Yingnan Xu

arxiv: 2501.12208 · v5 · pith:N6TJANQGnew · submitted 2025-01-21 · 💻 cs.SI

STEC-Net: A Spatiotemporal Graph Neural Framework for Community Discovery in Dynamic Social Networks

Yingnan Xu , Shuangshuang Chu This is my paper

Pith reviewed 2026-05-25 08:38 UTC · model grok-4.3

classification 💻 cs.SI

keywords community discoverydynamic social networksgraph convolutional networksgated recurrent unitsspatiotemporal embeddingsself-organizing mapsnetwork evolution

0 comments

The pith

STEC-Net discovers communities in dynamic networks by learning spatiotemporal embeddings from GCNs with evolving parameters, temporal GRUs, and SOM clustering.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces STEC-Net to improve community discovery in networks whose structure changes over time. Standard methods track only link formation and dissolution but overlook richer spatial patterns and how those patterns shift temporally. The framework applies graph convolutions to each network snapshot for spatial node features, uses a GRU to evolve the convolution weights as the topology changes, then applies another GRU across time steps to encode temporal dependencies into the embeddings. A self-organizing map finally clusters those embeddings into communities. Tests on four types of dynamic networks report higher scores on purity, normalized mutual information, homogeneity, and completeness than conventional approaches.

Core claim

STEC-Net integrates spatial structure and temporal dynamics within a unified embedding architecture. Graph Convolutional Networks learn snapshot-level node representations from network topology. A GRU-based weight evolution mechanism updates the GCN parameters over time to adapt to structural evolution. A second GRU models temporal dependencies across snapshot embeddings to produce spatiotemporal node representations. A Self-Organizing Map applied to the learned embeddings clusters nodes and infers their community affiliations, yielding consistent gains over traditional methods on dynamic network benchmarks.

What carries the argument

The STEC-Net pipeline that couples GCN spatial encoding with GRU-driven parameter evolution and a second temporal GRU before SOM clustering to generate adaptive spatiotemporal node embeddings.

If this is right

Community detection accuracy improves when spatial encoders adapt their weights to network evolution rather than using fixed parameters.
Temporal dependencies across snapshots can be modeled separately from spatial structure to produce more informative node embeddings.
Self-organizing maps provide an effective final step for assigning nodes to communities once spatiotemporal representations are obtained.
Performance gains appear across multiple metric types including purity, normalized mutual information, homogeneity, and completeness.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same architecture could be tested on link prediction or anomaly detection tasks in the same dynamic networks to check whether the embeddings generalize beyond clustering.
Scaling the method to networks with millions of nodes would require checking whether the per-snapshot GCN and dual GRUs remain computationally feasible.
Replacing the SOM with a differentiable clustering layer might allow end-to-end training and potentially tighter integration of the temporal signals.

Load-bearing premise

The specific combination of GCN spatial encoding, GRU-based parameter evolution, temporal GRU, and SOM clustering will reliably capture richer spatial-temporal structure than existing methods.

What would settle it

An ablation experiment on the same four network types in which removing the GRU weight evolution or the temporal GRU produces equal or higher purity and NMI scores would falsify the claim that the full combination is required.

read the original abstract

Community discovery is a central problem in the analysis of dynamic social networks. Traditional community discovery methods mainly focus on the formation and dissolution of links between nodes, and therefore often fail to capture the richer spatial structure and temporal dependency underlying network evolution. To address this limitation, we propose STEC-Net, a spatiotemporal graph neural framework for community discovery in dynamic social networks. STEC-Net integrates spatial structure and temporal dynamics within a unified embedding architecture. First, Graph Convolutional Networks (GCNs) are used to learn snapshot-level node representations from network topology. To adapt the spatial encoder to structural evolution, a GRU-based weight evolution mechanism is introduced to update the GCN parameters over time. Then, a second Gated Recurrent Unit (GRU) is employed to model temporal dependencies across snapshot embeddings and to learn spatiotemporal node representations. Finally, a Self-Organizing Map (SOM) is applied to the learned embeddings to cluster nodes and infer their community affiliations. Experiments on four types of dynamic networks show that STEC-Net consistently outperforms traditional community discovery methods in terms of purity, normalized mutual information, homogeneity, and completeness. These results demonstrate that STEC-Net can effectively uncover evolving community structures in dynamic social networks.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

STEC-Net wires GCNs to two GRUs and a SOM for dynamic community detection, but the outperformance claim has no visible experimental backing or controls.

read the letter

The one thing to know is that this paper describes a new pipeline—GCN spatial encoding whose weights evolve via GRU, followed by a temporal GRU and SOM clustering—but supplies zero experimental details to support the claim that it beats prior methods on purity, NMI, homogeneity, and completeness across four network types. The abstract is all we have, so the central result cannot be checked yet. The stress-test note lands cleanly here: without ablations isolating the GRU weight-evolution step, the second GRU, or SOM versus simpler clustering on the same embeddings, any gains could come from tuning or dataset choice rather than the claimed spatiotemporal integration. The paper does not show derivations or controls that would make the performance delta traceable to the architecture itself. What is actually new is the concrete combination of GRU-driven GCN parameter adaptation with a second temporal GRU and SOM output. That is a reasonable engineering extension of existing pieces rather than a first-principles advance, but it does target a genuine gap: most static community methods ignore how topology and node states change together over time. The write-up of that gap is straightforward and the high-level architecture description is easy to follow. The soft spot is entirely in the evaluation. No dataset descriptions, baseline lists, error bars, or component-wise results appear, so the soundness rating stays low. This is for people already working on dynamic social networks who might want to test a new GNN stack once the experiments are filled in. A reader who needs reproducible numbers or wants to cite a verified improvement will not find enough here. I would not send it to peer review until the authors add the missing methods, results, and ablations; the idea is clear enough that a revision could make it worth referee time.

Referee Report

1 major / 0 minor

Summary. The paper proposes STEC-Net, a framework that applies GCNs to learn node representations from network snapshots, uses a GRU-based mechanism to evolve GCN weights over time to adapt to structural changes, employs a second GRU to capture temporal dependencies across snapshot embeddings, and applies SOM clustering on the resulting spatiotemporal embeddings to discover communities. It reports that experiments on four types of dynamic networks demonstrate consistent outperformance over traditional methods on purity, normalized mutual information, homogeneity, and completeness.

Significance. If the reported gains are shown to arise specifically from the spatiotemporal integration rather than from tuning or individual components, the work could offer a practical architecture for dynamic community detection that jointly models topology evolution and temporal node trajectories, extending snapshot-based GCN approaches with explicit parameter adaptation and recurrent temporal modeling.

major comments (1)

[Experiments] Experiments section (and abstract): The central claim of consistent outperformance on four metrics across four network types rests on the premise that the full combination of GCN spatial encoding, GRU weight evolution, temporal GRU, and SOM is required to capture richer structure. No ablation studies, component-wise comparisons (e.g., static GCN vs. evolved weights, temporal GRU vs. concatenation, SOM vs. k-means on the same embeddings), or controls for hyper-parameter effects are described, leaving open the possibility that gains are driven by one submodule or dataset-specific tuning rather than the claimed integration.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback and for recognizing the potential of the spatiotemporal integration in STEC-Net. We address the single major comment below.

read point-by-point responses

Referee: [Experiments] Experiments section (and abstract): The central claim of consistent outperformance on four metrics across four network types rests on the premise that the full combination of GCN spatial encoding, GRU weight evolution, temporal GRU, and SOM is required to capture richer structure. No ablation studies, component-wise comparisons (e.g., static GCN vs. evolved weights, temporal GRU vs. concatenation, SOM vs. k-means on the same embeddings), or controls for hyper-parameter effects are described, leaving open the possibility that gains are driven by one submodule or dataset-specific tuning rather than the claimed integration.

Authors: We agree that the absence of ablation studies and component-wise comparisons leaves the source of the reported gains open to alternative explanations. The current experiments demonstrate that STEC-Net outperforms the listed traditional baselines on the four metrics across the four network types, but do not isolate the contribution of each architectural element. In the revision we will add the requested ablations (static GCN vs. GRU-evolved weights; temporal GRU vs. simple concatenation or no temporal modeling; SOM vs. k-means on identical embeddings) together with hyper-parameter sensitivity controls. These results will be reported in an expanded Experiments section and will be used to qualify the claims in the abstract and conclusion. revision: yes

Circularity Check

0 steps flagged

No circularity: architecture is independent of claimed performance

full rationale

The paper proposes STEC-Net as an explicit architectural stack (GCN spatial encoding + GRU weight evolution + temporal GRU + SOM clustering) and reports empirical outperformance on purity/NMI/homogeneity/completeness. No equations, derivations, or fitted-parameter predictions appear in the abstract or method description that would reduce the performance claims to quantities defined by the inputs themselves. No self-citations, uniqueness theorems, or ansatzes from prior author work are invoked to justify the design. The experimental results therefore constitute external validation rather than a self-referential reduction.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The framework rests on standard assumptions about the representational power of GCNs on static snapshots and GRUs on sequential data; no new free parameters, axioms, or invented entities are explicitly introduced or fitted in the abstract.

axioms (2)

domain assumption Graph Convolutional Networks produce useful node representations from network topology
Invoked as the base spatial encoder in the first stage.
domain assumption Gated Recurrent Units can model temporal dependencies across snapshot embeddings
Used both for weight evolution and for learning spatiotemporal representations.

pith-pipeline@v0.9.0 · 5745 in / 1202 out tokens · 28775 ms · 2026-05-25T08:38:36.999471+00:00 · methodology

Review history (2 revisions) →

discussion (0)

Reference graph

Works this paper leans on

32 extracted references · 32 canonical work pages

[1]

Computer Integrated Manufacturing System30(6), 2130– 2138 (2024)

Kang, H., Jing, W., Zhang, Y.: Overlapping community discovery algorithm based on industrial big data. Computer Integrated Manufacturing System30(6), 2130– 2138 (2024)

work page 2024
[2]

User Modeling and User-Adapted Interaction22(2012) https://doi.org/10.100 7/s11257-011-9111-y

Paliouras, G.: Discovery of web user communities and their role in personalization. User Modeling and User-Adapted Interaction22(2012) https://doi.org/10.100 7/s11257-011-9111-y

work page 2012
[3]

International Journal of Epidemiology33(3), 499–506 (2004) https://doi.org/10 .1093/ije/dyh010

Leung, M.W., Yen, I.H., Minkler, M.: Community based participatory research: a promising approach for increasing epidemiology’s relevance in the 21st century. International Journal of Epidemiology33(3), 499–506 (2004) https://doi.org/10 .1093/ije/dyh010

work page 2004
[4]

IEEE AccessPP, 1–1 (2023) https://doi.org/10.1109/ACCESS.2023.3260652

Rostami, M., Oussalah, M., Berahmand, K., Farrahi, V.: Community detection algorithms in healthcare applications: A systematic review. IEEE AccessPP, 1–1 (2023) https://doi.org/10.1109/ACCESS.2023.3260652

work page doi:10.1109/access.2023.3260652 2023
[5]

In: 2014 IEEE Security and Privacy Workshops, pp

Sarvari, H., Abozinadah, E., Mbaziira, A., Mccoy, D.: Constructing and analyzing criminal networks. In: 2014 IEEE Security and Privacy Workshops, pp. 84–91 (2014). https://doi.org/10.1109/SPW.2014.22

work page doi:10.1109/spw.2014.22 2014
[6]

In: SAS SUGI Proceedings: Customer Intelligence

Andr´ e, C., Pinheiro, R., De Janeiro, R.: Community detection to identify fraud events in telecommunications networks. In: SAS SUGI Proceedings: Customer Intelligence. SAS Global Forum 2012 (2012)

work page 2012
[7]

In: Pro- ceedings on the International Conference on Artificial Intelligence (ICAI), p

Waskiewicz, T.: Friend of a friend influence in terrorist social networks. In: Pro- ceedings on the International Conference on Artificial Intelligence (ICAI), p. 1 (2012). The Steering Committee of The World Congress in Computer Science, Computer

work page 2012
[8]

In: Mobile Radio Com- munications and 5G Networks: Proceedings of MRCN 2020, pp

Dhawan, S., Singh, K., Batra, A.: Community discovery and behavior prediction 22 in online social networks employing node centrality. In: Mobile Radio Com- munications and 5G Networks: Proceedings of MRCN 2020, pp. 73–85 (2021). Springer

work page 2020
[9]

IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans36(1), 208–219 (2006) https://doi.org/10.1109/TSMCA.2005.859095

Lancieri, L., Durand, N.: Internet user behavior: compared study of the access traces and application to the discovery of communities. IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans36(1), 208–219 (2006) https://doi.org/10.1109/TSMCA.2005.859095

work page doi:10.1109/tsmca.2005.859095 2006
[10]

Mini-Micro Systems45(4), 960–967 (2024)

Zhang, X., Wu, F., Jia, H., al.: Local community discovery algorithm that balances similarity and closeness. Mini-Micro Systems45(4), 960–967 (2024)

work page 2024
[11]

Acta Electronica Sinica52(8), 2786–2798 (2024)

He, C., Cheng, Q., Cheng, J., al.: A dynamic community discovery method that integrates node change information. Acta Electronica Sinica52(8), 2786–2798 (2024)

work page 2024
[12]

Computer Engineering and Applications60(21), 55–72 (2024)

Wu, Y., Dou, S., Li, J., al.: A review of community discovery research integrating heterogeneity and dynamics. Computer Engineering and Applications60(21), 55–72 (2024)

work page 2024
[13]

Library and Information Work64(9), 142–152 (2020)

Zhang, H., Zhou, H., Zhang, X., al.: Progress in community discovery research on online social networks. Library and Information Work64(9), 142–152 (2020)

work page 2020
[14]

Computer Science and Exploration16(1), 59–87 (2022)

Yuan, L., Li, X., Wang, X., al.: Overview of graph embedding models. Computer Science and Exploration16(1), 59–87 (2022)

work page 2022
[15]

Computer Science and Exploration15(4), 612–630 (2021)

Duan, X., Yuan, G., Meng, F.: A review of research on dynamic community discovery methods. Computer Science and Exploration15(4), 612–630 (2021)

work page 2021
[16]

Computer Application43(10), 3129–3135 (2023)

Yang, Y., Duan, W.: Dynamic community discovery algorithm for social networks based on spectral clustering. Computer Application43(10), 3129–3135 (2023)

work page 2023
[17]

Journal of Harbin University of Science and Technology28(3), 10–19 (2023)

Yang, H., Zhao, X., Chen, C., al.: Social network community discovery algorithm based on node influence expansion. Journal of Harbin University of Science and Technology28(3), 10–19 (2023)

work page 2023
[18]

Journal of Shandong University of Science and Technology (Natural Science Edition)40(4), 94–102 (2021)

Li, J., Cao, N., Zhang, Q., al.: Online social network group discovery algorithm considering theme and time. Journal of Shandong University of Science and Technology (Natural Science Edition)40(4), 94–102 (2021)

work page 2021
[19]

Physical Review E78(4) (2008) https://doi.org/10 .1103/physreve.78.046110

Lancichinetti, A., Fortunato, S., Radicchi, F.: Benchmark graphs for testing com- munity detection algorithms. Physical Review E78(4) (2008) https://doi.org/10 .1103/physreve.78.046110

work page 2008
[20]

In: Proceedings of the ACM SIGKDD Workshop on Mining Data Semantics, pp

Yang, J., Leskovec, J.: Defining and evaluating network communities based on ground-truth. In: Proceedings of the ACM SIGKDD Workshop on Mining Data Semantics, pp. 1–8 (2012) 23

work page 2012
[21]

In: The Political Communication Reader, pp

Cornfield, M.,et al.: Buzz, blogs, and beyond: The internet and the national discourse in the fall of 2004. In: The Political Communication Reader, pp. 296–

work page 2004
[22]

Routledge, ??? (2023)

work page 2023
[23]

IEEE Access13, 149071–149086 (2025) https://doi.org/10.1109/ACCESS.2025.3601649

Patel, B., Wilson, T.W., Stephen, J.M., Calhoun, V.D., Wang, Y.-P.: Mjnmf- gat: Multi-task joint non-negative matrix factorization graph attention network for understanding adolescent neurodevelopment. IEEE Access13, 149071–149086 (2025) https://doi.org/10.1109/ACCESS.2025.3601649

work page doi:10.1109/access.2025.3601649 2025
[24]

In: Proceedings of the 22nd International Conference on Knowledge Discovery and Data Mining, pp

Wang, D., Cui, P., Zhu, W.: Structural deep network embedding. In: Proceedings of the 22nd International Conference on Knowledge Discovery and Data Mining, pp. 1225–1234. ACM, New York (2023)

work page 2023
[25]

In: Proceedings of the 20th International Conference on Knowledge Discovery and Data Mining, pp

Perozzi, B., Al-Rfou, R., Skiena, S.: Deepwalk: Online learning of social repre- sentations. In: Proceedings of the 20th International Conference on Knowledge Discovery and Data Mining, pp. 701–710. ACM, New York (2014)

work page 2014
[26]

Knowledge-Based Systems18(7), 104816–104823 (2020)

Goyal, P., Chhetri, S.R., Canedo, A.: Dyngraph2vec: Capturing network dynamics using dynamic graph representation learning. Knowledge-Based Systems18(7), 104816–104823 (2020)

work page 2020
[27]

In: Proceedings of the 32th International Conference on World Wide Web, pp

Lin, Y.R., Chi, Y., Zhu, S.,et al.: Facetnet: A framework for analyzing com- munities and their evolutions in dynamic networks. In: Proceedings of the 32th International Conference on World Wide Web, pp. 685–694. ACM, New York (2023)

work page 2023
[28]

Journal of Statistical Mechanics: Theory and Experiment 21(10), 10008–10020 (2023)

Blondel, V.D., Guillaume, J.L., Lambiotte, R.,et al.: Fast unfolding of communi- ties in large networks. Journal of Statistical Mechanics: Theory and Experiment 21(10), 10008–10020 (2023)

work page 2023
[29]

In: Proceedings of the AAAI Conference on Artificial Intelligence, vol

Guo, S., Lin, Y., Feng, N., Song, C., Wan, H.: Attention based spatial-temporal graph convolutional networks for traffic flow forecasting. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 33, pp. 922–929 (2019)

work page 2019
[30]

IEEE Transactions on Neural Systems and Rehabilitation Engineering29, 1977–1986 (2021)

Jia, Z., Lin, Y., Wang, J., Ning, X., He, Y., Zhou, R., Zhou, Y., Lehman, L.H.: Multi-view spatial-temporal graph convolutional networks with domain general- ization for sleep stage classification. IEEE Transactions on Neural Systems and Rehabilitation Engineering29, 1977–1986 (2021)

work page 1977
[31]

In: Proceedings of the 31st ACM International Conference on Information & Knowledge Management (2022)

Song, J.,et al.: St-gat: A spatio-temporal graph attention network for accurate traffic speed prediction. In: Proceedings of the 31st ACM International Conference on Information & Knowledge Management (2022)

work page 2022
[32]

Engineering Applications of Artificial Intelligence132, 107947 (2024) 24

Huang, D., Song, J., He, Y.: Community detection algorithm for social network based on node intimacy and graph embedding model. Engineering Applications of Artificial Intelligence132, 107947 (2024) 24

work page 2024

[1] [1]

Computer Integrated Manufacturing System30(6), 2130– 2138 (2024)

Kang, H., Jing, W., Zhang, Y.: Overlapping community discovery algorithm based on industrial big data. Computer Integrated Manufacturing System30(6), 2130– 2138 (2024)

work page 2024

[2] [2]

User Modeling and User-Adapted Interaction22(2012) https://doi.org/10.100 7/s11257-011-9111-y

Paliouras, G.: Discovery of web user communities and their role in personalization. User Modeling and User-Adapted Interaction22(2012) https://doi.org/10.100 7/s11257-011-9111-y

work page 2012

[3] [3]

International Journal of Epidemiology33(3), 499–506 (2004) https://doi.org/10 .1093/ije/dyh010

Leung, M.W., Yen, I.H., Minkler, M.: Community based participatory research: a promising approach for increasing epidemiology’s relevance in the 21st century. International Journal of Epidemiology33(3), 499–506 (2004) https://doi.org/10 .1093/ije/dyh010

work page 2004

[4] [4]

IEEE AccessPP, 1–1 (2023) https://doi.org/10.1109/ACCESS.2023.3260652

Rostami, M., Oussalah, M., Berahmand, K., Farrahi, V.: Community detection algorithms in healthcare applications: A systematic review. IEEE AccessPP, 1–1 (2023) https://doi.org/10.1109/ACCESS.2023.3260652

work page doi:10.1109/access.2023.3260652 2023

[5] [5]

In: 2014 IEEE Security and Privacy Workshops, pp

Sarvari, H., Abozinadah, E., Mbaziira, A., Mccoy, D.: Constructing and analyzing criminal networks. In: 2014 IEEE Security and Privacy Workshops, pp. 84–91 (2014). https://doi.org/10.1109/SPW.2014.22

work page doi:10.1109/spw.2014.22 2014

[6] [6]

In: SAS SUGI Proceedings: Customer Intelligence

Andr´ e, C., Pinheiro, R., De Janeiro, R.: Community detection to identify fraud events in telecommunications networks. In: SAS SUGI Proceedings: Customer Intelligence. SAS Global Forum 2012 (2012)

work page 2012

[7] [7]

In: Pro- ceedings on the International Conference on Artificial Intelligence (ICAI), p

Waskiewicz, T.: Friend of a friend influence in terrorist social networks. In: Pro- ceedings on the International Conference on Artificial Intelligence (ICAI), p. 1 (2012). The Steering Committee of The World Congress in Computer Science, Computer

work page 2012

[8] [8]

In: Mobile Radio Com- munications and 5G Networks: Proceedings of MRCN 2020, pp

Dhawan, S., Singh, K., Batra, A.: Community discovery and behavior prediction 22 in online social networks employing node centrality. In: Mobile Radio Com- munications and 5G Networks: Proceedings of MRCN 2020, pp. 73–85 (2021). Springer

work page 2020

[9] [9]

IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans36(1), 208–219 (2006) https://doi.org/10.1109/TSMCA.2005.859095

Lancieri, L., Durand, N.: Internet user behavior: compared study of the access traces and application to the discovery of communities. IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans36(1), 208–219 (2006) https://doi.org/10.1109/TSMCA.2005.859095

work page doi:10.1109/tsmca.2005.859095 2006

[10] [10]

Mini-Micro Systems45(4), 960–967 (2024)

Zhang, X., Wu, F., Jia, H., al.: Local community discovery algorithm that balances similarity and closeness. Mini-Micro Systems45(4), 960–967 (2024)

work page 2024

[11] [11]

Acta Electronica Sinica52(8), 2786–2798 (2024)

He, C., Cheng, Q., Cheng, J., al.: A dynamic community discovery method that integrates node change information. Acta Electronica Sinica52(8), 2786–2798 (2024)

work page 2024

[12] [12]

Computer Engineering and Applications60(21), 55–72 (2024)

Wu, Y., Dou, S., Li, J., al.: A review of community discovery research integrating heterogeneity and dynamics. Computer Engineering and Applications60(21), 55–72 (2024)

work page 2024

[13] [13]

Library and Information Work64(9), 142–152 (2020)

Zhang, H., Zhou, H., Zhang, X., al.: Progress in community discovery research on online social networks. Library and Information Work64(9), 142–152 (2020)

work page 2020

[14] [14]

Computer Science and Exploration16(1), 59–87 (2022)

Yuan, L., Li, X., Wang, X., al.: Overview of graph embedding models. Computer Science and Exploration16(1), 59–87 (2022)

work page 2022

[15] [15]

Computer Science and Exploration15(4), 612–630 (2021)

Duan, X., Yuan, G., Meng, F.: A review of research on dynamic community discovery methods. Computer Science and Exploration15(4), 612–630 (2021)

work page 2021

[16] [16]

Computer Application43(10), 3129–3135 (2023)

Yang, Y., Duan, W.: Dynamic community discovery algorithm for social networks based on spectral clustering. Computer Application43(10), 3129–3135 (2023)

work page 2023

[17] [17]

Journal of Harbin University of Science and Technology28(3), 10–19 (2023)

Yang, H., Zhao, X., Chen, C., al.: Social network community discovery algorithm based on node influence expansion. Journal of Harbin University of Science and Technology28(3), 10–19 (2023)

work page 2023

[18] [18]

Journal of Shandong University of Science and Technology (Natural Science Edition)40(4), 94–102 (2021)

Li, J., Cao, N., Zhang, Q., al.: Online social network group discovery algorithm considering theme and time. Journal of Shandong University of Science and Technology (Natural Science Edition)40(4), 94–102 (2021)

work page 2021

[19] [19]

Physical Review E78(4) (2008) https://doi.org/10 .1103/physreve.78.046110

Lancichinetti, A., Fortunato, S., Radicchi, F.: Benchmark graphs for testing com- munity detection algorithms. Physical Review E78(4) (2008) https://doi.org/10 .1103/physreve.78.046110

work page 2008

[20] [20]

In: Proceedings of the ACM SIGKDD Workshop on Mining Data Semantics, pp

Yang, J., Leskovec, J.: Defining and evaluating network communities based on ground-truth. In: Proceedings of the ACM SIGKDD Workshop on Mining Data Semantics, pp. 1–8 (2012) 23

work page 2012

[21] [21]

In: The Political Communication Reader, pp

Cornfield, M.,et al.: Buzz, blogs, and beyond: The internet and the national discourse in the fall of 2004. In: The Political Communication Reader, pp. 296–

work page 2004

[22] [22]

Routledge, ??? (2023)

work page 2023

[23] [23]

IEEE Access13, 149071–149086 (2025) https://doi.org/10.1109/ACCESS.2025.3601649

Patel, B., Wilson, T.W., Stephen, J.M., Calhoun, V.D., Wang, Y.-P.: Mjnmf- gat: Multi-task joint non-negative matrix factorization graph attention network for understanding adolescent neurodevelopment. IEEE Access13, 149071–149086 (2025) https://doi.org/10.1109/ACCESS.2025.3601649

work page doi:10.1109/access.2025.3601649 2025

[24] [24]

In: Proceedings of the 22nd International Conference on Knowledge Discovery and Data Mining, pp

Wang, D., Cui, P., Zhu, W.: Structural deep network embedding. In: Proceedings of the 22nd International Conference on Knowledge Discovery and Data Mining, pp. 1225–1234. ACM, New York (2023)

work page 2023

[25] [25]

In: Proceedings of the 20th International Conference on Knowledge Discovery and Data Mining, pp

Perozzi, B., Al-Rfou, R., Skiena, S.: Deepwalk: Online learning of social repre- sentations. In: Proceedings of the 20th International Conference on Knowledge Discovery and Data Mining, pp. 701–710. ACM, New York (2014)

work page 2014

[26] [26]

Knowledge-Based Systems18(7), 104816–104823 (2020)

Goyal, P., Chhetri, S.R., Canedo, A.: Dyngraph2vec: Capturing network dynamics using dynamic graph representation learning. Knowledge-Based Systems18(7), 104816–104823 (2020)

work page 2020

[27] [27]

In: Proceedings of the 32th International Conference on World Wide Web, pp

Lin, Y.R., Chi, Y., Zhu, S.,et al.: Facetnet: A framework for analyzing com- munities and their evolutions in dynamic networks. In: Proceedings of the 32th International Conference on World Wide Web, pp. 685–694. ACM, New York (2023)

work page 2023

[28] [28]

Journal of Statistical Mechanics: Theory and Experiment 21(10), 10008–10020 (2023)

Blondel, V.D., Guillaume, J.L., Lambiotte, R.,et al.: Fast unfolding of communi- ties in large networks. Journal of Statistical Mechanics: Theory and Experiment 21(10), 10008–10020 (2023)

work page 2023

[29] [29]

In: Proceedings of the AAAI Conference on Artificial Intelligence, vol

Guo, S., Lin, Y., Feng, N., Song, C., Wan, H.: Attention based spatial-temporal graph convolutional networks for traffic flow forecasting. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 33, pp. 922–929 (2019)

work page 2019

[30] [30]

IEEE Transactions on Neural Systems and Rehabilitation Engineering29, 1977–1986 (2021)

Jia, Z., Lin, Y., Wang, J., Ning, X., He, Y., Zhou, R., Zhou, Y., Lehman, L.H.: Multi-view spatial-temporal graph convolutional networks with domain general- ization for sleep stage classification. IEEE Transactions on Neural Systems and Rehabilitation Engineering29, 1977–1986 (2021)

work page 1977

[31] [31]

In: Proceedings of the 31st ACM International Conference on Information & Knowledge Management (2022)

Song, J.,et al.: St-gat: A spatio-temporal graph attention network for accurate traffic speed prediction. In: Proceedings of the 31st ACM International Conference on Information & Knowledge Management (2022)

work page 2022

[32] [32]

Engineering Applications of Artificial Intelligence132, 107947 (2024) 24

Huang, D., Song, J., He, Y.: Community detection algorithm for social network based on node intimacy and graph embedding model. Engineering Applications of Artificial Intelligence132, 107947 (2024) 24

work page 2024