Extending Proximity Measures to Attributed Networks for Community Detection

2021 ◽  
Vol 30 (4) ◽  
pp. 441-455
Author(s):  
Rinat Aynulin ◽  
◽  
Pavel Chebotarev ◽  
◽  
Keyword(s):  
Network Analysis ◽  
Community Detection ◽  
Network Structure ◽  
Real World ◽  
Adjacency Matrix ◽  
Attribute Information ◽  
Node Attributes ◽  
Proximity Measures ◽  
Definition Of ◽  
Attributed Networks

Proximity measures on graphs are extensively used for solving various problems in network analysis, including community detection. Previous studies have considered proximity measures mainly for networks without attributes. However, attribute information, node attributes in particular, allows a more in-depth exploration of the network structure. This paper extends the definition of a number of proximity measures to the case of attributed networks. To take node attributes into account, attribute similarity is embedded into the adjacency matrix. Obtained attribute-aware proximity measures are numerically studied in the context of community detection in real-world networks.

Community detection via closure extension

2018 ◽  
Vol 29 (12) ◽  
pp. 1850119
Author(s):  
Jingming Zhang ◽  
Jianjun Cheng ◽  
Xiaosu Feng ◽  
Xiaoyun Chen
Keyword(s):  
Community Structure ◽  
Computational Complexity ◽  
Community Detection ◽  
Network Structure ◽  
Real World ◽  
Prior Information ◽  
State Of The Art ◽  
Local Information ◽  
Second Step ◽  
Novel Method

Identifying community structure in networks plays an important role in understanding the network structure and analyzing the network features. Many state-of-the-art algorithms have been proposed to identify the community structure in networks. In this paper, we propose a novel method based on closure extension; it performs in two steps. The first step uses the similarity closure or correlation closure to find the initial community structure. In the second step, we merge the initial communities using Modularity [Formula: see text]. The proposed method does not need any prior information such as the number or sizes of communities, and it is able to obtain the same resulting communities in multiple runs. Moreover, it is noteworthy that our method has low computational complexity because of considering only local information of network. Some real-world and synthetic graphs are used to test the performance of the proposed method. The results demonstrate that our method can detect deterministic and informative community structure in most cases.

scholarly journals Simulating systematic bias in attributed social networks and its effect on rankings of minority nodes

2021 ◽  
Vol 6 (1) ◽  
Author(s):  
Leonie Neuhäuser ◽  
Felix I. Stamm ◽  
Florian Lemmerich ◽  
Michael T. Schaub ◽  
Markus Strohmaier
Keyword(s):  
Network Analysis ◽  
Measurement Errors ◽  
Social Systems ◽  
Systematic Bias ◽  
Model Framework ◽  
Network Analyses ◽  
External Node ◽  
Systematic Biases ◽  
Node Attributes ◽  
Attributed Networks

AbstractNetwork analysis provides powerful tools to learn about a variety of social systems. However, most analyses implicitly assume that the considered relational data is error-free, and reliable and accurately reflects the system to be analysed. Especially if the network consists of multiple groups (e.g., genders, races), this assumption conflicts with a range of systematic biases, measurement errors and other inaccuracies that are well documented in the literature. To investigate the effects of such errors we introduce a framework for simulating systematic bias in attributed networks. Our framework enables us to model erroneous edge observations that are driven by external node attributes or errors arising from the (hidden) network structure itself. We exemplify how systematic inaccuracies distort conclusions drawn from network analyses on the task of minority representations in degree-based rankings. By analysing synthetic and real networks with varying homophily levels and group sizes, we find that the effect of introducing systematic edge errors depends on both the type of edge error and the level of homophily in the system: in heterophilic networks, minority representations in rankings are very sensitive to the type of systematic edge error. In contrast, in homophilic networks we find that minorities are at a disadvantage regardless of the type of error present. We thus conclude that the implications of systematic bias in edge data depend on an interplay between network topology and type of systematic error. This emphasises the need for an error model framework as developed here, which provides a first step towards studying the effects of systematic edge-uncertainty for various network analysis tasks.

A signal-diffusion-based spectral clustering method for community detection

2019 ◽  
Vol 18 (01) ◽  
pp. 1941019
Author(s):  
Zheng Qiong
Keyword(s):  
Community Detection ◽  
Real World ◽  
Adjacency Matrix ◽  
Spectral Clustering ◽  
Time Complexity ◽  
Detection Method ◽  
Signal Transmission ◽  
Transmission Mechanism ◽  
Clustering Method ◽  
Spectral Clustering Method

As the traditional spectral community detection method uses adjacency matrix for clustering which might cause the problem of accuracy reduction, we proposed a signal-diffusion-based spectral clustering for community detection. This method solves the problem that unfixed total signal as using the signal transmission mechanism, provides optimization of algorithm time complexity, improves the performance of spectral clustering with construction of Laplacian based on signal diffusion. Experiments prove that the method reaches as better performance on real-world network and Lancichinetti–Fortunato–Radicchi (LFR) benchmark.

scholarly journals NMLPA: Uncovering Overlapping Communities in Attributed Networks via a Multi-Label Propagation Approach

Sensors ◽  
2019 ◽  
Vol 19 (2) ◽  
pp. 260 ◽  
Author(s):  
Bingyang Huang ◽  
Chaokun Wang ◽  
Binbin Wang
Keyword(s):  
Community Detection ◽  
Real World ◽  
State Of The Art ◽  
Label Propagation ◽  
Detection Methods ◽  
Overlapping Communities ◽  
Propagation Algorithm ◽  
Pruning Strategy ◽  
Node Similarity ◽  
Attributed Networks

With the enrichment of the entity information in the real world, many networks with attributed nodes are proposed and studied widely. Community detection in these attributed networks is an essential task that aims to find groups where the intra-nodes are much more densely connected than the inter-nodes. However, many existing community detection methods in attributed networks do not distinguish overlapping communities from non-overlapping communities when designing algorithms. In this paper, we propose a novel and accurate algorithm called Node-similarity-based Multi-Label Propagation Algorithm (NMLPA) for detecting overlapping communities in attributed networks. NMLPA first calculates the similarity between nodes and then propagates multiple labels based on the network structure and the node similarity. Moreover, NMLPA uses a pruning strategy to keep the number of labels per node within a suitable range. Extensive experiments conducted on both synthetic and real-world networks show that our new method significantly outperforms state-of-the-art methods.

Hierarchical community detection with applications to real-world network analysis

2013 ◽  
Vol 83 ◽  
pp. 20-38 ◽  
Author(s):  
Bo Yang ◽  
Jin Di ◽  
Jiming Liu ◽  
Dayou Liu
Keyword(s):  
Network Analysis ◽  
Community Detection ◽  
Real World

Exploiting higher-order patterns for community detection in attributed graphs

2020 ◽  
pp. 1-12
Author(s):  
Lun Hu ◽  
Xiangyu Pan ◽  
Hong Yan ◽  
Pengwei Hu ◽  
Tiantian He
Keyword(s):  
Community Detection ◽  
Higher Order ◽  
Challenging Problem ◽  
Network Systems ◽  
Attributed Graphs ◽  
Attribute Information ◽  
Node Attributes ◽  
Real World Datasets ◽  
Different Characteristics ◽  
Do So

As a fundamental task in cluster analysis, community detection is crucial for the understanding of complex network systems in many disciplines such as biology and sociology. Recently, due to the increase in the richness and variety of attribute information associated with individual nodes, detecting communities in attributed graphs becomes a more challenging problem. Most existing works focus on the similarity between pairwise nodes in terms of both structural and attribute information while ignoring the higher-order patterns involving more than two nodes. In this paper, we explore the possibility of making use of higher-order information in attributed graphs to detect communities. To do so, we first compose tensors to specifically model the higher-order patterns of interest from the aspects of network structures and node attributes, and then propose a novel algorithm to capture these patterns for community detection. Extensive experiments on several real-world datasets with varying sizes and different characteristics of attribute information demonstrated the promising performance of our algorithm.

scholarly journals The item network and domain network of burnout in Chinese nurses

BMC Nursing ◽  
2021 ◽  
Vol 20 (1) ◽  
Author(s):  
Lin Wu ◽  
Lei Ren ◽  
Yifei Wang ◽  
Kan Zhang ◽  
Peng Fang ◽  
...  
Keyword(s):  
Network Analysis ◽  
Community Detection ◽  
Emotional Exhaustion ◽  
Network Structure ◽  
Partial Correlation ◽  
Clinical Intervention ◽  
Maslach Burnout Inventory ◽  
Cluster Sampling ◽  
Occupational Burnout ◽  
Nurse Burnout

Abstract Background As a common social phenomenon, nurses’ occupational burnout has a high incidence rate, which seriously affects their mental health and nursing level. The current assessment mostly uses the total score model and explores the influence of external factors on burnout, while the correlation between burnout items or dimensions is less explored. Ignoring the correlation between the items or dimensions may result in a limited understanding of nurse occupational burnout. This paper explores the item and dimension network structure of the Maslach Burnout Inventory-General Survey (MBI-GS) in Chinese nurses, so as to gain a deeper understanding of this psychological construct and identify potential targets for clinical intervention. Methods A total of 493 Chinese nurses were recruited by cluster sampling. All participants were invited to complete the survey on symptoms of burnout. Network analysis was used to investigate the item network of MBI-GS. In addition, community detection was used to explore the communities of MBI-GS, and then network analysis was used to investigate the dimension network of MBI-GS based on the results of community detection. Regularized partial correlation and non-regularized partial correlation were used to describe the association between different nodes of the item network and dimension network, respectively. Expected influence and predictability were used to describe the relative importance and the controllability of nodes in both the item and dimension networks. Results In the item network, most of the strongly correlated edges were in the same dimension of emotional exhaustion (E), cynicism (C) and reduced professional efficacy (R), respectively. E5 (Item 5 of emotional exhaustion, the same below) “I feel burned out from my work”, C1 “I have become more callous toward work since I took this job”, and R3 “In my opinion, I am good at my job” had the highest expected influence (z-scores = 0.99, 0.81 and 0.94, respectively), indicating theirs highest importance in the network. E1 “I feel emotionally drained from my work” and E5 had the highest predictability (E1 = 0.74, E5 = 0.74). It shows that these two nodes can be interpreted by their internal neighbors to the greatest extent and have the highest controllability in the network. The spinglass algorithm and walktrap algorithm obtained exactly the same three communities, which are consistent with the original dimensions of MBI-GS. In the dimension network, the emotional exhaustion dimension was closely related to the cynicism dimension (weight = 0.65). Conclusions The network model is a useful tool to study burnout in Chinese nurses. This study explores the item and domain network structure of nurse burnout from the network perspective. By calculating the relevant indicators, we found that E5, C1, and R3 were the most central nodes in the item network and cynicism was the central node in the domain network, suggesting that interventions aimed at E5, C1, R3 and cynicism might decrease the overall burnout level of Chinese nurses to the greatest extent. This study provides potential targets and a new way of thinking for the intervention of nurse burnout, which can be explored and verified in clinical practice.

scholarly journals Heterogeneous Attributed Network Embedding with Graph Convolutional Networks

2019 ◽  
Vol 33 ◽  
pp. 10061-10062 ◽  
Author(s):  
Yueyang Wang ◽  
Ziheng Duan ◽  
Binbing Liao ◽  
Fei Wu ◽  
Yueting Zhuang
Keyword(s):  
Real World ◽  
Network Embedding ◽  
High Quality ◽  
Embedding Method ◽  
The Real ◽  
Convolutional Networks ◽  
Attributed Network ◽  
Node Attributes ◽  
Attributed Networks

Network embedding which assigns nodes in networks to lowdimensional representations has received increasing attention in recent years. However, most existing approaches, especially the spectral-based methods, only consider the attributes in homogeneous networks. They are weak for heterogeneous attributed networks that involve different node types as well as rich node attributes and are common in real-world scenarios. In this paper, we propose HANE, a novel network embedding method based on Graph Convolutional Networks, that leverages both the heterogeneity and the node attributes to generate high-quality embeddings. The experiments on the real-world dataset show the effectiveness of our method.

Neighbors-based divisive algorithm for hierarchical analysis in networks

2019 ◽  
Vol 30 (07) ◽  
pp. 1940003
Author(s):  
Junhai Luo ◽  
Lei Ye ◽  
Xiaoting He
Keyword(s):  
Community Structure ◽  
Network Analysis ◽  
Hierarchical Clustering ◽  
Network Structure ◽  
Real World ◽  
Clustering Algorithms ◽  
Hierarchical Analysis ◽  
Multilevel Structure ◽  
Common Technique ◽  
The Common

Hierarchical analysis for network structure can point out which communities can constitute a larger group or give reasonable smaller groups within a community. Numerous methods for discovering community in networks divide networks at only one certain granularity, which does not benefit hierarchical analysis for network structure. Hierarchical clustering algorithms are the common technique that reveals the multilevel structure in the network analysis. In this work, we give a definition for scores of edges according to the basic idea of means clustering. Based on the definition, a neighbors-based divisive algorithm named neighbor-means (NM) is proposed to detect communities in networks, especially for hierarchical analysis. The divisive algorithm repeatedly removes the edge with the highest score to obtain hierarchical partitions but can recalculate the scores of edges quickly with local recalculating strategy and crucial change-rules, which makes its complexity much lower than many divisive algorithms. In addition, when the community structure is ambiguous, benefited from superiority of the defined scores, our method achieves better results than many divisive and agglomerative algorithms. Experiments with artificial and real-world networks demonstrate the superiority of neighbor-means in detecting community structure.

scholarly journals Neighborhood-based bridge node centrality tuple for complex network analysis

2021 ◽  
Vol 6 (1) ◽  
Author(s):  
Natarajan Meghanathan
Keyword(s):  
Network Analysis ◽  
Community Detection ◽  
Real World ◽  
Detection Algorithm ◽  
Algebraic Connectivity ◽  
Neighborhood Graph ◽  
Node Centrality ◽  
Centrality Metrics ◽  
Community Detection Algorithm ◽  
Topological Position

AbstractWe define a bridge node to be a node whose neighbor nodes are sparsely connected to each other and are likely to be part of different components if the node is removed from the network. We propose a computationally light neighborhood-based bridge node centrality (NBNC) tuple that could be used to identify the bridge nodes of a network as well as rank the nodes in a network on the basis of their topological position to function as bridge nodes. The NBNC tuple for a node is asynchronously computed on the basis of the neighborhood graph of the node that comprises of the neighbors of the node as vertices and the links connecting the neighbors as edges. The NBNC tuple for a node has three entries: the number of components in the neighborhood graph of the node, the algebraic connectivity ratio of the neighborhood graph of the node and the number of neighbors of the node. We analyze a suite of 60 complex real-world networks and evaluate the computational lightness, effectiveness, efficiency/accuracy and uniqueness of the NBNC tuple vis-a-vis the existing bridgeness related centrality metrics and the Louvain community detection algorithm.

Sign in / Sign up

Export Citation Format

Share Document