Article
Multidisciplinary Sciences
Huan Qing
Summary: A degree-corrected distribution-free model is proposed for weighted social networks with latent structural information. The model extends the previous distribution-free models and classical degree-corrected stochastic block model to fit real-world weighted networks. An algorithm based on spectral clustering is designed to fit the model, and theoretical framework and analysis are provided. The effectiveness of the proposed method is demonstrated through experiments.
SCIENTIFIC REPORTS
(2022)
Article
Computer Science, Artificial Intelligence
Furkan Oztemiz, Ali Karci
Summary: This study proposes a modularity optimization algorithm to increase clustering success in any network without being dependent on any community detection algorithm. The algorithm transfers nodes at the community boundary to neighboring communities to improve the modularity value. Experimental results show that the proposed method significantly enhances the modularity values of community detection algorithms.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Operations Research & Management Science
Alessandro Chessa, Pierpaolo D'Urso, Livia De Giovanni, Vincenzina Vitale, Alfonso Gebbia
Summary: This paper uses a weighted complex network and a sparsification procedure to detect communities of basketball players. It calculates the best community structure and maximizes modularity as a measure of compactness and separation. The effectiveness of the sparsification transition is confirmed. This method not only finds the best distribution of nodes and number of communities, but also enables a data-driven decision-making process in basketball.
ANNALS OF OPERATIONS RESEARCH
(2023)
Article
Mathematics, Interdisciplinary Applications
Wenyi Fang, Xin Wang, Longzhao Liu, Zhaole Wu, Shaoting Tang, Zhiming Zheng
Summary: This paper proposes a gradient descent framework called vector-label propagation algorithm (VLPA) for modularity optimization in community detection. By retaining weak structural information in vector-label, VLPA outperforms some well-known community detection methods, especially in networks with weak community structures. The authors further incorporate stochastic gradient strategies into VLPA and develop the stochastic vector-label propagation algorithm (sVLPA), which performs better than the widely used Louvain Method on artificial benchmarks and real-world networks.
CHAOS SOLITONS & FRACTALS
(2022)
Article
Computer Science, Software Engineering
Youcef Belkhiri, Nadjet Kamel, Habiba Drias
Summary: This article introduces a multi-swarm elephant herding optimization algorithm for detecting hiding communities in complex networks. By updating clan procedure and separating procedure, the algorithm aims to uncover community structures through interacting clans and determining the best local individuals.
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
(2022)
Article
Multidisciplinary Sciences
Kai Qi, Heng Zhang, Yang Zhou, Yifan Liu, Qingxiang Li
Summary: This study introduces an algorithm called PR-LFM, which combines an improved local fitness maximization (LFM) algorithm with the PageRank (PR) algorithm for community partitioning on cyberspace resources. The experimental data demonstrate good results in the resource division of cyberspace.
SCIENTIFIC REPORTS
(2023)
Article
Computer Science, Artificial Intelligence
Antonio G. Spampinato, Rocco A. Scollo, Vincenzo Cutello, Mario Pavone
Summary: Community detection, an important research topic in Complex Network Analysis, plays a significant role in interpreting and understanding various systems in neuroscience, biology, social science, and economy. This paper introduces an immune optimization algorithm (opt-IA) for detecting community structures, aiming to maximize the modularity of the identified communities. Compared with 20 heuristics and metaheuristics, opt-IA demonstrates superior performance while being comparable to the Hyper-Heuristic method. The results confirm that opt-IA, despite relying on a purely random process, is reliable and efficient.
Article
Physics, Multidisciplinary
Guolin Wu, Changgui Gu, Huijie Yang
Summary: In this study, we found that the method of modularity and the method of graph partitioning are essentially equivalent in bipartite networks. We also proposed a spectral algorithm of modularity for identifying community structures, which outperforms other modularity algorithms in synthetic and real-world networks.
Article
Biodiversity Conservation
Siti Aisyah Tumiran, Bellie Sivakumar
Summary: Catchment classification is important for various studies, and the concept of community structure plays a significant role. The proposed MDEB method shows better performance compared to the EB method in practice.
ECOLOGICAL INDICATORS
(2021)
Article
Physics, Multidisciplinary
Jing Xiao, Xiao-ke Xu
Summary: This review provides an overview of two emerging research directions in complex network analysis - fuzzy and higher-order community detection. It covers related concepts, mathematical descriptions, latest advancements, current challenges, and future directions.
Article
Operations Research & Management Science
Haibin Chen, Hongjin He, Yiju Wang, Guanglu Zhou
Summary: This paper discusses a class of NP-hard fourth degree polynomial problems, focusing on the equivalence between Bi-QOP and MOP over compact sets with concave objective functions, introducing an augmented Bi-QOP to maintain concavity, and proposing an algorithm to find approximate optimal values. Computational results on synthetic datasets demonstrate the efficiency of the algorithm.
JOURNAL OF GLOBAL OPTIMIZATION
(2022)
Article
Construction & Building Technology
Zhixiang Hu, Huiyu Zhu, Lei Huang, Cheng Cheng
Summary: A two-stage damage identification method for beam structures based on support vector machine and swarm intelligence optimization algorithms is proposed in this paper. The method first obtains the frequencies and mode shapes of the beam structure using the smooth orthogonal decomposition method, and then calculates the normalized modal curvature as the input of a pre-trained support vector machine to determine the damage location. The stiffness loss at the damaged location of the structure is then calculated using swarm intelligence algorithms.
Article
Computer Science, Artificial Intelligence
Rouhollah Javadpour Boroujeni, Seyfollah Soleimani
Summary: Community detection is a method to simplify network analysis and understand network behavior and function. Modularity is a common measure used in various approaches, and identifying influential nodes is also important for community detection in complex networks.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Mathematics
Bo Zhang, Yifei Mi, Lele Zhang, Yuping Zhang, Maozhen Li, Qianqian Zhai, Meizi Li
Summary: This study proposes an incremental dynamic community detection model based on a graph neural network node embedding representation. By improving the information enrichment of node feature vectors and introducing a new modularity calculation method, it can detect dynamic communities more accurately.
Article
Multidisciplinary Sciences
Keyou Guo, Chengbo He, Min Yang, Sudong Wang
Summary: This study utilizes an improved YOLOv5 model for automatic detection and recognition of pavement distresses, and introduces attention mechanism to enhance the robustness of the model. Experimental results show that the improved model can effectively identify pavement distresses on an intelligent mobile platform.
SCIENTIFIC REPORTS
(2022)