Co-Association Matrix-Based Multi-Layer Fusion for Community Detection in Attributed Networks

Overlapping Community Detection in Vehicular Social Networks Based on Graph Attention Autoencoder

Community detection is particularly important in vehicular social networks because it helps identify closely connected groups of vehicles within the network. Community structures with overlapping relationships are identified through network topology and vehicle attribute information, thereby optimizing communication efficiency, supporting resource allocation, and enhancing privacy protection. However, most existing community detection methods focus on non-overlapping communities, usually only considering the topological structure of the network, and often ignoring the attribute information of nodes. To address these problems, this paper proposes a semi-supervised overlapping community detection method based on graph attention autoencoder (CDGAAE). The method consists of three key components: graph attention autoencoder module, modularity optimization enhancement module, and semi-supervised clustering module. First, the graph attention autoencoder module fuses topological information and node attribute information and encodes nodes using a graph attention mechanism. Second, the modularity optimization enhancement module effectively captures the structure of overlapping communities. Finally, the semi-supervised clustering module combines prior information to improve the accuracy of community detection. CDGAAE is comprehensively evaluated on multiple real and synthetic datasets, and experimental results show that CDGAAE outperforms other competing methods.

F-Deepwalk: A Community Detection Model for Transport Networks

The design of transportation networks is generally performed on the basis of the division of a metropolitan region into communities. With the combination of the scale, population density, and travel characteristics of each community, the transportation routes and stations can be more precisely determined to meet the travel demand of residents within each of the communities as well as the transportation links among communities. To accurately divide urban communities, the original word vector sampling method is improved on the classic Deepwalk model, proposing a Random Walk (RW) algorithm in which the sampling is modified with the generalized travel cost and improved logit model. Urban spatial community detection is realized with the K-means algorithm, building the F-Deepwalk model. Using the basic road network as an example, the experimental results show that the Deepwalk model, which considers the generalized travel cost of residents, has a higher profile coefficient, and the performance of the model improves with the reduction of random walk length. At the same time, taking the Shijiazhuang urban rail transit network as an example, the accuracy of the model is further verified.

Improved brain community structure detection by two-step weighted modularity maximization

The human brain can be regarded as a complex network with interacting connections between brain regions. Complex brain network analyses have been widely applied to functional magnetic resonance imaging (fMRI) data and have revealed the existence of community structures in brain networks. The identification of communities may provide insight into understanding the topological functions of brain networks. Among various community detection methods, the modularity maximization (MM) method has the advantages of model conciseness, fast convergence and strong adaptability to large-scale networks and has been extended from single-layer networks to multilayer networks to investigate the community structure changes of brain networks. However, the problems of MM, suffering from instability and failing to detect hierarchical community structure in networks, largely limit the application of MM in the community detection of brain networks. In this study, we proposed the weighted modularity maximization (WMM) method by using the weight matrix to weight the adjacency matrix and improve the performance of MM. Moreover, we further proposed the two-step WMM method to detect the hierarchical community structures of networks by utilizing node attributes. The results of the synthetic networks without node attributes demonstrated that WMM showed better partition accuracy than both MM and robust MM and better stability than MM. The two-step WMM method showed better accuracy of community partitioning than WMM for synthetic networks with node attributes. Moreover, the results of resting state fMRI (rs-fMRI) data showed that two-step WMM had the advantage of detecting the hierarchical communities over WMM and was more insensitive to the density of the rs-fMRI networks than WMM.