Community Detection Algorithm Research Articles

Spark-Based Label Diffusion and Label Selection Community Detection Algorithm for Metagenome Sequence Clustering

It is a challenge to assemble an enormous amount of metagenome data in metagenomics. Usually, metagenome cluster sequence before assembly accelerates the whole process. In SpaRC, sequences are defined as nodes and clustered by a parallel label propagation algorithm (LPA). To address the randomness of label selection from the parallel LPA during clustering and improve the completeness of metagenome sequence clustering, Spark-based parallel label diffusion and label selection community detection algorithm is proposed in the paper to obtain more accurate clustering results. In this paper, the importance of sequence is defined based on the Jaccard similarity coefficient and its degree. The core sequence is defined as the one with the largest importance in its located community. Three strategies are formulated to reduce the randomness of label selection. Firstly, the core sequence label diffuses over its located cluster and becomes the initial label of other sequences. Those sequences that do not receive an initial label will select the sequence label with the highest importance in the neighbor sequences. Secondly, we perform improved label propagation in order of label frequency and sequence importance to reduce the randomness of label selection. Finally, a merge small communities step is added to increase the completeness of clustered clusters. The experimental results show that our proposed algorithm can effectively reduce the randomness of label selection, improve the purity, completeness, and F-Measure and reduce the runtime of metagenome sequence clustering.

Quantitative and qualitative interpretation of community partitions by map overlaying and calculating the distribution of related geographical features

Abstract Applying community detection algorithms in spatial interaction networks constructed from modern human communication records is an essential means of evaluating urban territorial subdivisions. Previous studies have usually involved qualitative rather than quantitative interpretations of community detection results. This article proposes a method of quantitatively and qualitatively interpreting community partition results by map overlaying the spatial regions corresponding to the detected communities with the related geographical features and by calculating the distribution of the geographical features contained in the regions and the entropy value of each distribution. The interpretation of the communities detected from the spatial interaction networks is carried out from the perspective of multi-temporal and multi-spatial scales and multi-geographical features. Extensive experiments were conducted with Milan, Italy, as the study area. The spatial interaction records reflected by telephone calls, land use, and point of interest (POI) data were used as the experimental data. Experimental results demonstrated the effectiveness of our method, and the specific results include: (1) Qualitative interpretation of multi-spatial resolution scale communities detected from the long-term aggregated spatial interaction network. The cohesiveness, homogeneity, and heterogeneity of the detected communities were qualitatively interpreted by the spatial distribution patterns of the land use dataset and the POI dataset. (2) Quantitative interpretation of multi-spatial resolution scale communities detected from the long-term aggregated spatial interaction network. The low spatial resolution scale community partitions and the high spatial resolution scale community partitions were interpreted through the statistical distribution of the land use dataset and the POI dataset, respectively. (3) Qualitative interpretation of the stable and active regions discovered from the community time series. Regardless of the community partitions’ spatial resolution scales, the stable and active regions were distinguished with the statistical distributions of the land use dataset and the POI dataset.

Spread and control of medical rumors in a social network: A generalized diffusion model with a highly asymmetric network structure

Medical rumors have become a threat to modern society. To study the spread and control of rumors, nonlinear differential equations modeling with the well‐mixed assumption is commonly used. However, this approach ignores the underlying network structure which plays an important role in information spreading. We establish a generalized differential equations model to study the spread and control of medical rumors in a highly asymmetric social network. In our model, each node represents a group of people and a “weighted” and “directed” network describes the communications between these nodes. This network can be generated from real‐world data by community detection algorithms. We provide methods to numerically calculate the final size of a rumor in each node and its derivatives with respect to each parameter. With these methods, if the government has resources to influence the parameters subject to certain constraints or cost functions, one can obtain the optimal resources allocation easily through nonlinear programming algorithms. We show that the implications on the government's resources allocation from the well‐mixed special case in the literature or conventional wisdom may become inapplicable in the general situation. Therefore, the underlying network should not be ignored. Because the final size of a medical rumor is not always the best measure of its damage, we extend our results to a wide class of objectives and show that different objectives result in very different implications. While the lack of a rule of thumb may sound negative, our flexible framework provides a powerful workhorse for interested parties to work out the details in their specific situations. Finally, we provide a sufficient condition for no outbreak of rumors. This condition can serve as a heuristic that a government with abundant resources can use to prevent the outbreak of rumors.

Robustness of community structure under edge addition.

Communities often represent key structural and functional clusters in networks. To preserve such communities, it is important to understand their robustness under network perturbations. Previous work in community robustness analysis has focused on studying changes in the community structure as a response of edge rewiring and node or edge removal. However, the impact of increasing connectivity on the robustness of communities in networked systems is relatively unexplored. Studying the limits of community robustness under edge addition is crucial to better understanding the cases in which density expands or false edges erroneously appear. In this paper, we analyze the effect of edge addition on community robustness in synthetic and empirical temporal networks. We study two scenarios of edge addition: random and targeted. We use four community detection algorithms, Infomap, Label Propagation, Leiden, and Louvain, and demonstrate the results in community similarity metrics. The experiments on synthetic networks show that communities are more robust when the initial partition is stronger or the edge addition is random, and the experiments on empirical data also indicate that robustness performance can be affected by the community similarity metric. Overall, our results suggest that the communities identified by the different types of community detection algorithms exhibit different levels of robustness, and so the robustness of communities depends strongly on the choice of detection method.

Overlapping community detection in software ecosystem based on pheromone guided personalized PageRank algorithm

Software ecosystem has aroused the interest of numerous researchers and plays an important role in many aspects. According to different participants and their relationships, software ecosystem can be constructed into various types of complex networks. Overlapping community detection in complex networks can help reveal the community structure and find the intersection between communities. However, existing overlapping community detection algorithms often suffer from reduced applicability or accuracy when applied to networks in software ecosystems. To reveal the overlapping community structure in software ecosystem, we propose an overlapping community detection algorithm by improving the standard personalized PageRank (PPR) algorithm. We first construct a developer collaboration network in software ecosystem based on the intensity of cooperation between developers. Then, the similarity between developers is calculated to guide the walking process of the PPR algorithm, making it suitable for weighted networks. Finally, in the proposed algorithm PGPPR, inspired by the idea of using pheromones to guide the walking in ant colony algorithm, we run the algorithm in multiple rounds and use the results of the previous round to guide the walking process in the current round to reduce redundant diffusion. The experimental results on the five real-world networks show that our algorithm is applicable and effective in detecting communities. And in the five developer collaboration networks, PGPPR can effectively detect overlapping community structures with higher stability and accuracy than the four baselines. The PGPPR algorithm can find overlapping community structures in weighted networks and effectively reduce the redundant diffusion generated when applying the standard PPR algorithm to community detection. Compared to other algorithms, our algorithm can detect the overlapping communities more accurately and stably when applied to developer collaboration networks in software ecosystem.

Dynamic community detection based on graph convolutional networks and contrastive learning

With the continuous development of technology and networks, real-life interactions are gradually being abstracted into social networks for study. Social circles are a fundamental structural feature that is prevalent on social networks. Thus, exploring social circle structure plays an important role in revealing the characteristics of complex social networks and provides guidance for understanding social behavior in real life. For example, it can aid in precision marketing, personalized recommendation, and knowledge dissemination within social circles. One of the important means of identifying social network circles lies on community detection algorithms. However, real-world social networks are often dynamic and can be studied and analyzed by building dynamic networks, while existing dynamic network community detection methods tend to ignore the global structure information and time-series information of nodes. To address this problem, this paper proposes a dynamic network community detection algorithm based on graph convolutional neural networks and contrastive learning, which fully captures the adjacent characteristics between nodes based on the correlation information and leverages the feature smoothing strategy to efficiently extract node representations of dynamic networks under unsupervised scenario. Specifically, the proposed algorithm first utilizes node correlation based aggregation strategy to compute the feature matrix for single time-step of the dynamic network. Then, mutual information maximization is implemented based on cross-entropy between learned local and global representations. To reduce the computational overhead in the optimization process, an additional LSTM module is further equipped for updating the parameters of graph convolutional networks in each time-step. Additionally, a contrastive learning based network smoothing strategy is designed to minimize the feature differences between neighboring nodes. Comparative experiments demonstrate that the proposed algorithm achieves excellent performance on both synthetic and real networks.

Evaluating Community Detection Algorithms for Multilayer Networks: Effectiveness of Link Weights and Link Direction

Analyzing the structures of multilayer networks (MLNs) has been a hot research topic in network science. Community detection algorithms are important tools for analyzing MLNs. In the literature, several community detection algorithms for MLNs have been proposed. Moreover, there are several options for the graph representation of an MLN: for example, directed or undirected, weighted or unweighted, and using information from all or only some layers. Although these options may affect the results of community detection in MLNs, representations that are effective for community detection have not yet been clarified. In this paper, we experimentally evaluate the effectiveness of three types of community detection algorithms for MLNs and examine how the graph representation of an MLN affects the results of these algorithms. Our main findings are as follows: (1) The flattening approach is particularly effective, whereas the layer-by-layer approach is not applicable to detecting communities in MLNs of Twitter users. (2) Using a directed graph for each layer of an MLN increases the accuracy of community detection. (3) The Leiden method, which is a community detection algorithm for single-layer networks, achieves comparable accuracy with the community detection algorithms for MLNs, which suggests that there exists room for improvement in multilayer community detection algorithms for effectively utilizing the multilayer structures of MLNs.

An adaptive dynamic community detection algorithm based on multi-objective evolutionary clustering

PurposeCommunity detection of dynamic networks provides more effective information than static network community detection in the real world. The mainstream method for community detection in dynamic networks is evolutionary clustering, which uses temporal smoothness of community structures to connect snapshots of networks in adjacent time intervals. However, the error accumulation issues limit the effectiveness of evolutionary clustering. While the multi-objective evolutionary approach can solve the issue of fixed settings of the two objective function weight parameters in the evolutionary clustering framework, the traditional multi-objective evolutionary approach lacks self-adaptability.Design/methodology/approachThis paper proposes a community detection algorithm that integrates evolutionary clustering and decomposition-based multi-objective optimization methods. In this approach, a benchmark correction procedure is added to the evolutionary clustering framework to prevent the division results from drifting.FindingsExperimental results demonstrate the superior accuracy of this method compared to similar algorithms in both real and synthetic dynamic datasets.Originality/valueTo enhance the clustering results, adaptive variances and crossover probabilities are designed based on the relative change amounts of the subproblems decomposed by MOEA/D (A Multiobjective Optimization Evolutionary Algorithm based on Decomposition) to dynamically adjust the focus of different evolutionary stages.

Modeling and exploring the evolution of the mobile software ecosystem: How far are we?

AbstractThe health of mobile software ecosystems is closely related to the interests of software developers, end‐users, and stakeholders. Therefore, it is crucial to maintain the mobile software ecosystem healthy and functioning. Researchers have done considerable research on mobile software ecosystems like Android and iOS. However, the evolution laws implicit in mobile software ecosystems have not attracted widespread attention. This paper proposes a research framework for investigating the evolution process and influencing factors of mobile software ecosystems based on community mining. Firstly, we mine the evolving ecosystem from many mobile software projects based on a community detection algorithm. Then we analyze the evolution process of the ecosystem by identifying evolution events in different periods. Furthermore, we utilize the multinomial logistics regression model to analyze the relevant indicators and summarize the crucial factors affecting the evolution. Meanwhile, by training the long short term memory (LSTM) model to predict evolution events, our prediction accuracy can reach 75%. This work can be used to maintain and improve the healthy operations of mobile software ecosystems.

Performance Analysis of a Multicore Approach Proposed for Efficient Community Detection and Recommendation System

Community detection is a well-known area of research that yields essential results in various fields such as social media, biological networks, pandemic spread, recommendation systems, etc. There are two important areas for improvement in existing community detection algorithms: quality of community formed should be improved and a parallel approach to community detection is needed to handle massive data. In this paper, we have proposed a three-step parallel algorithm, Par-Com, using the concept of k Clique and modularity optimization to address both of the above issues. The proposed algorithm increases the execution speed and also improves the quality of the community formed by optimizing modularity. Par-Com uses dynamic load balancing on the multicore architecture of Supercomputer ParamShivay. We have also evaluated Par-Com’s performance against nine sequential and three parallel community detection algorithms on varying size datasets, i.e. karate, macaque, email, immuno, soc-epinions, facebook, and com-Friendster. The experiment result shows that Par-Com outperforms other algorithms under consideration with up to 45% increase in modularity and up to 84% increase in execution speed. Par-Com is also capable of detecting overlapping communities, fuzzy membership of each node, most influential node in each community formed, and outlier nodes. Nodes within a community that have the highest influence are deemed as experts. The choices made by an expert in a particular community are served as recommendations to other users within that community.

An Exploratory Application of Multilayer Networks and Pathway Analysis in Pharmacogenomics.

Over the years, network analysis has become a promising strategy for analysing complex system, i.e., systems composed of a large number of interacting elements. In particular, multilayer networks have emerged as a powerful framework for modelling and analysing complex systems with multiple types of interactions. Network analysis can be applied to pharmacogenomics to gain insights into the interactions between genes, drugs, and diseases. By integrating network analysis techniques with pharmacogenomic data, the goal consists of uncovering complex relationships and identifying key genes to use in pathway enrichment analysis to figure out biological pathways involved in drug response and adverse reactions. In this study, we modelled omics, disease, and drug data together through multilayer network representation. Then, we mined the multilayer network with a community detection algorithm to obtain the top communities. After that, we used the identified list of genes from the communities to perform pathway enrichment analysis (PEA) to figure out the biological function affected by the selected genes. The results show that the genes forming the top community have multiple roles through different pathways.

SAT412 Effects Of Gender-Affirming Hormone Therapy On The Connectivity Of Intrinsic Brain Networks

Abstract Disclosure: C.C. Pedreira: None. S. Tchir-Bourgeois: None. S. Tuli: None. A. Giancola: None. C. Loxton: None. A.S. Keuroghlian: None. M. Misra: None. J.M. Andreano: None. Background: Previous research has indicated that both acute gonadal hormone treatments and fluctuations in ovarian hormone levels over the course of the menstrual cycle can significantly alter patterns of communication between brain regions, particularly regions involved in affective processing. However, the neural effects of prolonged exogenous gonadal hormone exposure, as in gender-affirming hormone therapy (GAHT), are not presently well-understood. A better understanding of the influence of GAHT on intrinsic network connectivity could point the way towards improved GAHT treatment regimens that minimize affective side effects.Objective: To characterize the effects of estradiol and testosterone treatments in GAH on patterns of connectivity within and between the brain’s intrinsic networks.Methods: 5 transgender individuals receiving GAHT (estradiol n =3, testosterone n = 2) were scanned using functional magnetic imaging (fMRI) at rest at two timepoints: baseline, prior to treatment, and 6 months into GAHT. The time course of neural activity was computed for each voxel in the brain, then tested for correlation with each other voxel to develop a whole-brain connectome for each participant. Connectomes were then analyzed in two ways: 1) maps of average connectivity to two regions of interest found in previous research to differ between cis and transgender people (the anterior insula and temporal pole) were generated for each participant. These maps were then compared between baseline and 6-month timepoints. 2) Overall network structure of the brain was generated a priori for each participant at each timepoint using a community detection algorithm. Network assignments were then compared between timepoints, noting regions whose connectivity shifted between networks after GAHT.Results: 2 of 3 participants treated with estradiol exhibited a significant reduction of connectivity to the anterior insula throughout the brain, most notably within the medial temporal lobe and cingulate cortex. Both participants receiving testosterone treatment showed reduced anterior insula connectivity, although connectivity to the cingulate cortex, particularly the mid-cingulate, was preserved. Estradiol did not significantly affect connectivity of the frontal pole, although testosterone produced a widespread reduction in frontal pole connectivity. Community detection analysis indicated a shift in connectivity to the salience network from the default mode network in the medial temporal lobe following estradiol treatment. Testosterone treatment was associated with greater connectivity in medial frontal regions to the frontoparietal network. Conclusions: Preliminary evidence indicates that both estradiol and testosterone treatment can produce significant changes in the connectivity of affective regions in the brain, as well as reorganization of the brain’s network structure. Presentation: Saturday, June 17, 2023

Re-evaluating the importance of threatened species in maintaining global phytoregions.

Anthropogenic introductions are known to be changing the structure of global phytogeographical regions (phytoregions), but previous studies have been limited by incomplete orbiased data sets that are likely to underestimate the importance of threatened species. In this work, we analyse a comprehensive data set of all known species and their occurrences (at botanical country resolution) to quantify the impact of potential future extinction scenarios. We used Infomap, a network-based community detection algorithm, to generate phytoregional delineations for six species-distribution scenarios (native, introduced and extinctions of species that are either documented as threatened or likely to be threatened, as well as combinations thereof). We compared the numbers and sizes of phytoregions to characterise the amount and spatial distribution of changes in global phytoregions under each scenario. Extinctions of species that are predicted to be threatened had a greater homogenising effect on phytoregions than introductions, and there was some evidence that introductions may even mitigate the homogenisation caused by extinctions, though this interaction is complex. This research provides the first evidence that the loss of threatened species would have significant ramifications for global phytoregions and demonstrates the need to consider extinction processes in studies of anthropogenic effects on biodiversity patterns.

A Data-Driven Rutting Depth Short-Time Prediction Model with Metaheuristic Optimization for Asphalt Pavements Based on RIOHTrack

Rutting of asphalt pavements is a crucial design criterion in various pavement design guides. A good road transportation base can provide security for the transportation of oil and gas in road transportation. This study attempts to develop a robust artificial intelligence model to estimate different asphalt pavements' rutting depth clips, temperature, and load axes as primary characteristics. The experiment data were obtained from 19 asphalt pavements with different crude oil sources on a 2.038 km long full-scale field accelerated pavement test track (RIOHTrack, Road Track Institute) in Tongzhou, Beijing. In addition, this paper also proposes to build complex networks with different pavement rutting depths through complex network methods and the Louvain algorithm for community detection. The most critical structural elements can be selected from different asphalt pavement rutting data, and similar structural elements can be found. An extreme learning machine algorithm with residual correction (RELM) is designed and optimized using an independent adaptive particle swarm algorithm. The experimental results of the proposed method are compared with several classical machine learning algorithms, with predictions of Average Root Mean Squared Error, Average Mean Absolute Error, and Average Mean Absolute Percentage Error for 19 asphalt pavements reaching 1.742, 1.363, and 1.94\% respectively. The experiments demonstrate that the RELM algorithm has an advantage over classical machine learning methods in dealing with non-linear problems in road engineering. Notably, the method ensures the adaptation of the simulated environment to different levels of abstraction through the cognitive analysis of the production environment parameters.

A Low-Memory Community Detection Algorithm With Hybrid Sparse Structure and Structural Information for Large-Scale Networks

A Low-Memory Community Detection Algorithm With Hybrid Sparse Structure and Structural Information for Large-Scale Networks

Community deception in directed influence networks

Community deception is about protecting users of a community from being discovered by community detection algorithms. This paper studies community deception in directed influence network (DIN). It aims to address the limitations of the state of the art through a twofold strategy: introducing directed influence and considering the role of nodes in the deception strategy. The study focuses on using modularity as the optimization function. It offers several contributions, including an upgraded version of modularity that accommodates the concept of influence, edge-based, and node-based deception algorithms.. The study concludes with a comparison of the proposed methods with the state of the art showing that not only influence is a valuable ingredient to devising deception strategies but also that novel deception approaches centered on node operations can be successfully devised.

Unsupervised Community Detection Algorithm with Stochastic Competitive Learning Incorporating Local Node Similarity

Community detection is an important task in the analysis of complex networks, which is significant for mining and analyzing the organization and function of networks. As an unsupervised learning algorithm based on the particle competition mechanism, stochastic competitive learning has been applied in the field of community detection in complex networks, but still has several limitations. In order to improve the stability and accuracy of stochastic competitive learning and solve the problem of community detection, we propose an unsupervised community detection algorithm LNSSCL (Local Node Similarity-Integrated Stochastic Competitive Learning). The algorithm calculates node degree as well as Salton similarity metrics to determine the starting position of particle walk; local node similarity is incorporated into the particle preferential walk rule; the particle is dynamically adjusted to control capability increments according to the control range; particles select the node with the strongest control capability within the node to be resurrected; and the LNSSCL algorithm introduces a node affiliation selection step to adjust the node community labels. Experimental comparisons with 12 representative community detection algorithms on real network datasets and synthetic networks show that the LNSSCL algorithm is overall better than other compared algorithms in terms of standardized mutual information (NMI) and modularity (Q). The improvement effect for the stochastic competition learning algorithm is evident, and it can effectively accomplish the community detection task in complex networks.

Community-Based Matrix Factorization (CBMF) Approach for Enhancing Quality of Recommendations.

Matrix factorization is a long-established method employed for analyzing and extracting valuable insight recommendations from complex networks containing user ratings. The execution time and computational resources demanded by these algorithms pose limitations when confronted with large datasets. Community detection algorithms play a crucial role in identifying groups and communities within intricate networks. To overcome the challenge of extensive computing resources with matrix factorization techniques, we present a novel framework that utilizes the inherent community information of the rating network. Our proposed approach, named Community-Based Matrix Factorization (CBMF), has the following steps: (1) Model the rating network as a complex bipartite network. (2) Divide the network into communities. (3) Extract the rating matrices pertaining only to those communities and apply MF on these matrices in parallel. (4) Merge the predicted rating matrices belonging to communities and evaluate the root mean square error (RMSE). In our experimentation, we use basic MF, SVD++, and FANMF for matrix factorization, and the Louvain algorithm is used for community division. The experimental evaluation on six datasets shows that the proposed CBMF enhances the quality of recommendations in each case. In the MovieLens 100K dataset, RMSE has been reduced to 0.21 from 1.26 using SVD++ by dividing the network into 25 communities. A similar reduction in RMSE is observed for the datasets of FilmTrust, Jester, Wikilens, Good Books, and Cell Phone.

A Survey of Malware Analysis Using Community Detection Algorithms

In recent years, we have witnessed an overwhelming and fast proliferation of different types of malware targeting organizations and individuals, which considerably increased the time required to detect malware. The malware developers make this issue worse by spreading many variants of the same malware [ 13 ]. To deal with this issue, graph theory techniques, and particularly community detection algorithms, can be leveraged to achieve bulk detection of malware families and variants to identify malicious communities instead of focusing on the detection of an individual instance of malware, which could significantly reduce the detection time. In this article, we review the state-of-the-art malware analysis solutions that employ community detection algorithms and provide a taxonomy that classifies the solutions with respect to five facets: analysis task, community detection approach, target platform, analysis type, and source of features. We present the solutions with respect to the analysis task, which covers malware detection, malware classification, cyber-threat infrastructure detection, and feature selection. The findings of this survey indicate that there is still room for contributions to further improve the state of the art and address research gaps. Finally, we discuss the advantages and the limitations of the solutions, identify open issues, and provide future research directions.

A novel approach to optimize flow capacity of the polder canal networks through community detection-based topology modification

Polder, a typical landscape in coastal areas, was facing threats from both environmental degradation and pluvial flood, as a result of the decreasing flow capacity in its canal networks which were continually paved during urban development. Filling or connecting canals was the major strategy adopted in such system that modifying its topological structure and was proved to be correlated with its overall flow capacity (r2 = 0.5488). Thus, we proposed a planning method to improve the flow capacity of polder canal systems which will be helpful in strategic water management. With the hypothesis of equivalent expected flow in loop canal networks, an optimum method based on the community-detection algorithm was formulated to partition the canals into groups and make connection plans. Further, a case study was carried out in Wusong Polder (53.5 km2), with 32.8 km canals out of 148 km (133 canals) being low-velocity canals (average velocity < 0.005 m/s, LVC). The current system was clustered into 21 to 22 groups depending on different starting nodes. The best performing plan reduced the total LVC length to 26.3 km by connecting 10 canals (6.09 km). The robustness of flow capacity improvements from the optimized plan and its contribution to flood control was validated.