Temporal Community Detection Research Articles

HB-DSBM: Modeling the Dynamic Complex Networks From Community Level to Node Level.

A variety of methods have been proposed for modeling and mining dynamic complex networks, in which the topological structure varies with time. As the most popular and successful network model, the stochastic block model (SBM) has been extended and applied to community detection, link prediction, anomaly detection, and evolution analysis of dynamic networks. However, all current models based on the SBM for modeling dynamic networks are designed at the community level, assuming that nodes in each community have the same dynamic behavior, which usually results in poor performance on temporal community detection and loses the modeling of node abnormal behavior. To solve the above-mentioned problem, this article proposes a hierarchical Bayesian dynamic SBM (HB-DSBM) for modeling the node-level and community-level dynamic behavior in a dynamic network synchronously. Based on the SBM, we introduce a hierarchical Dirichlet generative mechanism to associate the global community evolution with the microscopic transition behavior of nodes near-perfectly and generate the observed links across the dynamic networks. Meanwhile, an effective variational inference algorithm is developed and we can easy to infer the communities and dynamic behaviors of the nodes. Furthermore, with the two-level evolution behaviors, it can identify nodes or communities with abnormal behavior. Experiments on simulated and real-world networks demonstrate that HB-DSBM has achieved state-of-the-art performance on community detection and evolution. In addition, abnormal evolutionary behavior and events on dynamic networks can be effectively identified by our model.

Exploring Temporal Community Structure via Network Embedding.

Temporal community detection is helpful to discover and analyze significant groups or clusters hidden in dynamic networks in the real world. A variety of methods, such as modularity optimization, spectral method, and statistical network model, has been developed from diversified perspectives. Recently, network embedding-based technologies have made significant progress, and one can exploit deep learning superiority to network tasks. Although some methods for static networks have shown promising results in boosting community detection by integrating community embedding, they are not suitable for temporal networks and unable to capture their dynamics. Furthermore, the dynamic embedding methods only model network varying without considering community structures. Hence, in this article, we propose a novel unsupervised dynamic community detection model, which is based on network embedding and can effectively discover temporal communities and model dynamic networks. More specifically, we propose the community prior by introducing the Gaussian mixture model (GMM) in the variational autoencoder, which can obtain community information and better model the evolutionary characteristics of community structure and node embedding by utilizing the variant of gated recurrent unit (GRU). Extensive experiments conducted in real-world and artificial networks demonstrate that our proposed model has a better effect on improving the accuracy of dynamic community detection.

Between news and history: identifying networked topics of collective attention on Wikipedia

The digital information landscape has introduced a new dimension to understanding how we collectively react to new information and preserve it at the societal level. This, together with the emergence of platforms such as Wikipedia, has challenged traditional views on the relationship between current events and historical accounts of events, with an ever-shrinking divide between “news” and “history”. Wikipedia’s place as the Internet’s primary reference work thus poses the question of how it represents both traditional encyclopaedic knowledge and evolving important news stories. In other words, how is information on and attention towards current events integrated into the existing topical structures of Wikipedia? To address this, we develop a temporal community detection approach towards topic detection that takes into account both short term dynamics of attention as well as long term article network structures. We apply this method to a dataset of one year of current events on Wikipedia to identify clusters of Wikipedia articles related to news events, distinct from those that would be found solely from page view time series correlations or static network structure. We are able to resolve the topics that more strongly reflect unfolding current events vs more established knowledge by the relative importance of collective attention dynamics vs link structures. We also offer important developments by identifying and describing the emergent topics on Wikipedia. This work provides a means of distinguishing how these information and attention clusters are related to Wikipedia’s twin faces of encyclopaedic knowledge and current events—crucial to understanding the production and consumption of knowledge in the digital age.

Modeling Community Evolution Characteristics of Dynamic Networks with Evolutionary Bayesian Nonnegative Matrix Factorization

In most cases, the block structures and evolution characteristics always coexist in dynamic networks. This leads to inaccurate results of temporal community structure analysis with a two-step strategy. Fortunately, a few approaches take the evolution characteristics into account for modeling temporal community structures. But the number of communities cannot be determined automatically. Therefore, a model, Evolutionary Bayesian Nonnegative Matrix Factorization (EvoBNMF), is proposed in this paper. It focuses on modeling the temporal community structures with evolution characteristics. More specifically, the evolution behavior, which is introduced into EvoBNMF, can quantify the transfer intensity of communities between adjacent snapshots for modeling the evolution characteristics. Innovatively, the most appropriate number of communities can be determined autonomously by shrinking the corresponding evolution behaviors. Experimental results show that our approach has superior performance on temporal community detection with the virtue of autonomous determination of the number of communities.

Visibility graph based temporal community detection with applications in biological time series

Temporal behavior is an essential aspect of all biological systems. Time series have been previously represented as networks. Such representations must address two fundamental problems on how to: (1) Create appropriate networks to reflect the characteristics of biological time series. (2) Detect characteristic dynamic patterns or events as network temporal communities. General community detection methods use metrics comparing the connectivity within a community to random models, or are based on the betweenness centrality of edges or nodes. However, such methods were not designed for network representations of time series. We introduce a visibility-graph-based method to build networks from time series and detect temporal communities within these networks. To characterize unevenly sampled time series (typical of biological experiments), and simultaneously capture events associated to peaks and troughs, we introduce the Weighted Dual-Perspective Visibility Graph (WDPVG). To detect temporal communities in individual signals, we first find the shortest path of the network between start and end nodes, identifying high intensity nodes as the main stem of our community detection algorithm that act as hubs for each community. Then, we aggregate nodes outside the shortest path to the closest nodes found on the main stem based on the closest path length, thereby assigning every node to a temporal community based on proximity to the stem nodes/hubs. We demonstrate the validity and effectiveness of our method through simulation and biological applications.

Exploring the transition behavior of nodes in temporal networks based on dynamic community detection

Community detection and community evolution tracking are two important tasks in dynamic complex network analysis. Recently, a variety of models and methods have been proposed for detecting the community structure and analyzing their evolution. However, all these methods are only committed to improving the performance of community detection or identifying evolutionary events, ignoring the internal relevance between the structure of each snapshot of the dynamic network and the evolution pattern of communities, especially the structural features of nodes and their dynamic transition behavior. To cope with this problem, we firstly conduct experiments on 15 real-world dynamic networks to explore the transition behavior of nodes in dynamic networks, which is one of the most influential evolutionary patterns in temporal community detection. Firstly, we obtain the temporal community structure based on very successful temporal community detection methods. Secondly, we extract features of nodes based on the structure of the dynamic network, and take the community transition behavior of nodes as the binary classification problem. Finally, we use the decision tree to find the node-level features that have a general impact on node transition. Experiments indicate that the degree and average neighbor degree of nodes have the most common indispensable impact on the node transition behavior, which are very helpful for modeling dynamic complex networks in future.

Overlapping Community Detection in Weighted Temporal Text Networks

Network is a powerful language to represent relational data. One way to understand network is to analyze groups of nodes which share same properties or functions. The task of discovering such groups is known as community detection. The community detection in real-life networks, the majority of which are weighted temporal text networks, is confronted with two main problems - how to model the weight of edges and how to exploit the temporal information. Existing works either ignore the edge weight or utilize it in graph measures like modularity, which lacks scalability. And currently the common-used method involving temporal information is to discretize the time, which leads to series of problems. We are thus motivated to present a new method to encode the edge weight and temporal information. A probabilistic generative model, named Custom Temporal Community Detection (CTCD) is introduced, which views the link between two nodes as a weighted edge with several time stamps. Our model utilizes network, semantic and temporal information simultaneously to extract temporal community affiliations for individual user, influence strength across communities and temporal interested topic in each community. An efficient inference method, which scales linearly, and corresponding parallel implementation are proposed to adapt to large datasets. Through the knowledge extracted by CTCD, we are able to spot the community shift of the individual user, to which little attention has been given, and employ it to track the development of the communities over time. Moreover, experiments on two large-scale weighted temporal text networks show that CTCD gains significant improvement over state-of-the-art methods on a series of tasks.

Enhanced Community Detection Based on Cross Time for Higher Visibility In Supply Chain: A Six-Steps Model Framework

Increasing the visibility in supply chain network had decrease the risk in industries. However, the current Cross-Time approach for temporal community detection algorithm in the visibility has fix number of communities and lack of operation such as split or merge. Therefore, improving temporal community detection algorithm to represent the relationship in supply chain network for higher visibility is significant. This paper proposed six steps model framework that aim: (1) To construct the nodes and vertices for temporal graph representing the relationship in supply chain network; (2) To propose an enhanced temporal community detection algorithm in graph analytics based on Cross-time approach and (3) To evaluate the enhanced temporal community detection algorithm in graph analytics for representing relationship in supply chain network based on external and internal quality analysis. The proposed framework utilizes the Cross-Time approach for enhancing temporal community detection algorithm. The expected result shows that the Enhanced Temporal Community Detection Algorithm based on Cross Time approach for higher visibility in supply chain network is the major finding when implementing this proposed framework. The impact advances industrialization through efficient supply chain in industry leading to urbanization.

Temporally intermittent communities in brain fMRI correlation networks

It is widely agreed that the human brain is organized as a system of segregated modules that reside in separate regions and, through coordinated integration, support different cognitive functions. Through recent breakthroughs in modeling the activity of the brain, it has been demonstrated that each such module can participate in multiple so-called functional networks – networks of brain regions that activate in synchrony during specific types of cognition. If we model the brain as a temporal network, by representing brain regions as nodes and correlations within activity-windows at different times as links, we can formulate the task of finding functional networks as a community detection problem. In spite of this, however, relatively little attention has been given to solving this problem using recently developed techniques for temporal community detection. In this paper, as a proof-of-concept, we apply a novel technique for community detection in temporal networks to a dataset of fMRI measurements from 100 healthy subjects undertaking a working memory task with intermittent fixation (or resting-state) periods. We show that this method recovers two distinct communities that are shared between subjects: one that activates during the fixation period, and another that activates during a period associated with high cognitive load.

Temporal Label Walk for Community Detection and Tracking in Temporal Network

The problem of temporal community detection is discussed in this paper. Main existing methods are either structure-based or incremental analysis. The difficulty of the former is to select a suitable time window. The latter needs to know the initial structure of networks and the changing of networks should be stable. For most real data sets, these conditions hardly prevail. A streaming method called Temporal Label Walk (TLW) is proposed in this paper, where the aforementioned restrictions are eliminated. Modularity of the snapshots is used to evaluate our method. Experiments reveal the effectiveness of TLW on temporal community detection. Compared with other static methods in real data sets, our method keeps a higher modularity with the increase of window size.

A novel evolutionary clustering via the first-order varying information for dynamic networks

Temporal community detection could help us analyze and understand the meaningful substructure hidden within dynamic networks in the real world. Evolutionary clustering, as a popular framework for clustering stream data, has been denoted for mining the communities in dynamic networks. However, most of these methods ignore the varying characteristics of micro structure of the networks and lack of statistical interpretation. In this paper, we propose a powerful, interpretable and extensible evolutionary clustering framework based on nonnegative matrix factorization (NMF) for temporal community detection via combining the first-order varying information of micro structure in dynamic networks from the perspective of statistical model. Firstly, we consider the first-order varying information of nodes by constructing a temporal similarity matrix over time. Secondly, we present the framework, FVI-NMF, for detecting temporal community based on NMF combining the First-order Varying Information. Thirdly, we develop a effective algorithm to optimize the objective function of FVI-NMF and analyze its complexity. In addition, our model can discover the evolutionary pattern of temporal communities synchronously, which has a variety applications in the analysis of dynamic network. Experiments on both artificial and real dynamic networks demonstrate that our proposed framework has superior performance in comparison with state-of-art methods.

Modeling the Local and Global Evolution Pattern of Community Structures for Dynamic Networks Analysis

Exploring and understanding the temporal structure of dynamic networks attract extensive attention over the past few years. Most of these current research focuses on temporal community detection, evolution analysis or link prediction from a mission-oriented perspective. In fact, these three tasks should be not isolated but mutually reinforcing. Transforming these three tasks into a unified framework, it is crucial to extract the evolution pattern, which helps to understand the time-varying characteristics of temporal structure in essence. In addition, to the best of our knowledge, there is no work focusing on modeling and uncovering the local and global evolution pattern hidden in temporal community structure, simultaneously. In this paper, we propose a novel framework based on Orthogonal Nonnegative Matrix Factorization to Explore the Evolution Pattern (ONMF-EEP) for analyzing and predicting the time-varying structures in dynamic networks from local and global perspectives. The nature of this framework assumes that community structures are subject to a local evolution pattern (LEP) at each snapshot, and these LEPs are from a common global evolution pattern (GEP). The framework can synchronously detect temporal community structure, extract evolution pattern, and predict structure including communities and future snapshot links. The extensive experiments on real-world networks and artificial networks demonstrate that our proposed framework is highly effective on the tasks of dynamic network analysis.

Exploring temporal community structure and constant evolutionary pattern hiding in dynamic networks

Abstract Detecting the temporal communities and analyzing their evolution in dynamic networks is an important question for understanding the structure and function of complex networks. Most existing methods deal the temporal community detection and evolution as a two-step processes and only apply to unweighted and undirected dynamic networks. In this paper, we proposed a new clustering method based on non-negative matrix factorization from a fully probabilistic perspective, to explore temporal and constant community structure as well as the importance of nodes in any type dynamic networks synchronously. In detailed, we first denote the important matrix of node in dynamic networks, the community membership matrix, the similarity matrix at each snapshot and the probability transition matrix of community between the two consecutive snapshots. Second, we constitute the temporal community detection method from the view of generating networks. Third, we introduce a gradient descent algorithm to optimize the objection function of the proposed model. The experimental results on both artificial and real dynamic networks demonstrate that the superior performance of our proposed method is over some widely-used methods.

Community Detection in Temporal Networks Using Triple Nonnegative Matrix Factorization

Analyzing temporal networks can uncover dynamic evolution and characterize the properties of the networks. This paper proposes a novel temporal community detection model using triple nonnegative matrix factorization. Node weight matrices are introduced for targeting central nodes of communities and reducing number of nodes that have unobvious propensities of belonging to communities, which improves the algorithm performance of community detection. Community membership temporal smoothness constraint is added to discover latent structure and evolutionary behaviors of temporal networks. We then propose a gradient descent algorithm to optimize objective function. Experimental results on synthetic and real benchmarked networks show the effectiveness of detecting communities and finding their temporal changes.

Temporal community detection based on symmetric nonnegative matrix factorization

To understand time-evolving networks, researchers should not only concentrate on the community structures, an essential property of complex networks, in each snapshot, but also study the internal evolution of the entire networks. Temporal communities provide insights into such mechanism, i.e., how the communities emerge, expand, shrink, merge, split and decay over time. Based on the symmetric nonnegative matrix factorization (SNMF), we present a dynamic model to detect temporal communities, which not only could find a well community structure in a given snapshot but also demands the results bear some similarity to the partition obtained from the previous snapshot. Moreover, our method can handle the situation that of the number of community changes in the networks. Also, a gradient descent algorithm is proposed to optimize the objective function of the model. Experimental results on both the synthetic and real-world networks indicate that our method outperforms the state-of-art methods for temporal community detection.