Network Centrality Measures Research Articles

A network analysis of subjective well-being in Chinese high school students

BackgroundThe psychological situation of high school students during adolescence is not promising, and the most obvious manifestation is the lack of subjective well-being (SWB). This network analysis presents a model of the interaction and correlation between different items of SWB, identifying the most central items for high school students.MethodsThrough offline and online surveys, 4,378 questionnaires were sent out and finally 4,282 Chinese high school students were available. The response rate was 97.807%. The study used the eLASSO method to estimate the network structure and centrality measures. This algorithm used the EBIC to select the best neighbor factor for each node.ResultsThe average age for high school students was 16.320 years old and the average SWB score was 76.680. The distribution of SWB between male and female students was significant different (P < 0.001). S8 (Have you been anxious, worried, or upset) was the node with the highest strength and expected influence. The network structure and centrality remained stable after discarding 75% of the sample at random. Except for S15 (How concerned or worried about your health have you been), all nodes were positively correlated with each other (P < 0.01). The network structure of SWB was similar for female and male students (network strength: 8.482 for male participants; 8.323 for female participants; P = 0.159), as well as for rural and urban students (network strength: 8.500 for rural students; 8.315 for urban students; P = 0.140).ConclusionTargeting S8 (Have you been anxious, worried, or upset) as a potential intervention target may increase high school students’ SWB effectively.

Hypergraph models of the mental lexicon capture greater information than pairwise networks for predicting language learning

Human memory is a complex system that works in associative ways: Reading a cue word can lead to the recollection of associated concepts. The network structure of memory recall patterns has been shown to contain insights about a wide variety of cognitive phenomena, including language acquisition. However, most current network approaches use pairwise connections, i.e. links between only two words at a time. This ignores the possibility that more than two concept representations might be simultaneously associated in memory. We overcome this modelling limitation by introducing cognitive hypergraphs as models of human memory. We model memory recall patterns through word associations from the Small World of Words project for N=6003 concepts (Study 1) and for N=497 concepts (Study 2). In each study we represent word associations as either a pairwise network or a hypergraph. By combining psycholinguistic norms and network centrality measures with machine learning, we quantitatively investigate whether there is any benefit to using the hypergraph model over a pairwise network in predicting test-based age of acquisition norms in children up to age 9 years (Study 1) or normative learning in toddlers up to age 30 months (Study 2, based on CHILDES data). We show that cognitive hypergraphs capture more information than pairwise networks from the same data: Cognitive hypergraphs are considerably more powerful than pairwise networks at predicting age of acquisition trends in toddlers, children and teenagers. Our studies showcase how novel approaches merging artificial intelligence and higher-order interactions can help us understand cognitive development.

An Analysis of Key Actor Networks for Scale-Up Strategies for Childhood Obesity Prevention and the Care of Children with Obesity in Brazil

BackgroundEffective scale-up of multisectoral strategies aimed to prevent and treat childhood obesity has been a challenge in Brazil, the largest country in Latin America. Implementation Science methods, such as Net-Map, can identify key actors and opinion leaders (OLs) to advance the implementation and promote sustainability. ObjectivesThis study aimed to analyze power relations between key actors and OLs who influence the scale-up of Brazilian strategies for childhood obesity at the federal and state/municipal (local) levels. MethodsA mixed method study, applying the Net-Map method, collected data through virtual workshops with federal and local level stakeholders. The Net-Map included key actors mapping, power mapping, and identification of OLs. Four domains of power were analyzed: command, funding, technical assistance, and dissemination. Network cohesion and centrality measures were calculated. A qualitative analysis was conducted to qualify power relations according to ∗ gears for a successful scale-up (i.e., coordination, goals, and monitoring; advocacy; political will; legislation and policy; funding and resources; training; program delivery; communication; and research and technical cooperation). ResultsA total of 121 federal key actors and 63 local key actors were identified across networks, of which 62 and 28 were identified as OLs, respectively. Whereas the command domain of power had the highest number of key actors, the funding domain had the least. The health sector executive branch emerged as an OL across all domains of power. ConclusionsBarriers that threatened successful scale-up include the lack of coordination between domains of power, missing leadership within key actors, and lack of mechanisms to manage conflict of interest. Governance strategies to enhance multisectoral coordination and communication are needed to effectively scale-up and sustain childhood obesity strategies in Brazil.

Debiased personalized gene coexpression networks for population-scale scRNA-seq data.

Population-scale single-cell RNA-seq (scRNA-seq) data sets create unique opportunities for quantifying expression variation across individuals at the gene coexpression network level. Estimation of coexpression networks is well established for bulk RNA-seq; however, single-cell measurements pose novel challenges owing to technical limitations and noise levels of this technology. Gene-gene correlation estimates from scRNA-seq tend to be severely biased toward zero for genes with low and sparse expression. Here, we present Dozer to debias gene-gene correlation estimates from scRNA-seq data sets and accurately quantify network-level variation across individuals. Dozer corrects correlation estimates in the general Poisson measurement model and provides a metric to quantify genes measured with high noise. Computational experiments establish that Dozer estimates are robust to mean expression levels of the genes and the sequencing depths of the data sets. Compared with alternatives, Dozer results in fewer false-positive edges in the coexpression networks, yields more accurate estimates of network centrality measures and modules, and improves the faithfulness of networks estimated from separate batches of the data sets. We showcase unique analyses enabled by Dozer in two population-scale scRNA-seq applications. Coexpression network-based centrality analysis of multiple differentiating human induced pluripotent stem cell (iPSC) lines yields biologically coherent gene groups that are associated with iPSC differentiation efficiency. Application with population-scale scRNA-seq of oligodendrocytes from postmortem human tissues of Alzheimer's disease and controls uniquely reveals coexpression modules of innate immune response with distinct coexpression levels between the diagnoses. Dozer represents an important advance in estimating personalized coexpression networks from scRNA-seq data.

Branching processes reveal influential nodes in social networks

Branching processes are discrete-time stochastic processes which have been largely employed to model and simulate information diffusion processes over large online social networks such as Twitter and Reddit. Here we show that a variant of the branching process model enables the prediction of the popularity of user-generated content and thus can serve as a method for ranking search results or suggestions displayed to users. The proposed branching-process variant is able to evaluate the importance of an agent in a social network and, thus we propose a novel centrality index, called the Stochastic Potential Gain (SPG). The SPG is the first centrality index which combines the knowledge of the network topology with a dynamic process taking place on it which we call a graph-driven branching process. SPG generalises a range of popular network centrality metrics such as Katz' and Subgraph. We formulate a Monte Carlo algorithm (called MCPG) to compute the SPG and prove that it is convergent and correct. Experiments on two real datasets drawn from Facebook and GitHub demonstrate that MCPG traverses only a small fraction of nodes to produce its result, thus making the Stochastic Potential Gain an appealing option to compute node centrality measure for Online social networks.

Competitive pricing under local network effects

This paper will study the pricing problem of two competitive products in a market characterized by local externalities. For this purpose, a stochastic model of sales propagation among consumers is developed. This model utilizes a compartmentalized schema denoted as a Markov Chain where the local network effects impact transition rates. A key aspect of the proposed model is its multilayer structure, where products’ information streams through different layers. The equilibrium conditions and the optimal pricing strategies in a competitive market are examined. The pricing problem is investigated in two settings of homogeneous and heterogeneous (i.e., differential). It is shown that the existence and coexistence of individual products in the equilibrium point depends on an epidemic parameter, called reproduction number, that quantifies the speed by which a product’s sales spread over the network. Moreover, it is found that the correlation between the network’s layers impacts the equilibrium point. Specifically, a negative correlation between the network’s layers allows a wider coexistence region than a positive correlation. Additionally, it is found that a negative correlation between the network’s layers provides more flexibility to firms for their pricing practices and yields a higher profit. Finally, different pricing strategies are characterized with respect to model parameters and the centrality measures of different networks. It is observed that while centrality measures and optimal prices are highly correlated, node centralities alone are not enough to determine optimal prices.

Centrality measures in networks

Centrality measures in networks

MultiCens: Multilayer network centrality measures to uncover molecular mediators of tissue-tissue communication.

With the evolution of multicellularity, communication among cells in different tissues and organs became pivotal to life. Molecular basis of such communication has long been studied, but genome-wide screens for genes and other biomolecules mediating tissue-tissue signaling are lacking. To systematically identify inter-tissue mediators, we present a novel computational approach MultiCens (Multilayer/Multi-tissue network Centrality measures). Unlike single-layer network methods, MultiCens can distinguish within- vs. across-layer connectivity to quantify the "influence" of any gene in a tissue on a query set of genes of interest in another tissue. MultiCens enjoys theoretical guarantees on convergence and decomposability, and performs well on synthetic benchmarks. On human multi-tissue datasets, MultiCens predicts known and novel genes linked to hormones. MultiCens further reveals shifts in gene network architecture among four brain regions in Alzheimer's disease. MultiCens-prioritized hypotheses from these two diverse applications, and potential future ones like "Multi-tissue-expanded Gene Ontology" analysis, can enable whole-body yet molecular-level systems investigations in humans.

Network Structure of Comorbidity Patterns in U.S. Adults with Depression: A National Study Based on Data from the Behavioral Risk Factor Surveillance System.

People with depression are at increased risk for comorbidities; however, the clustering of comorbidity patterns in these patients is still unclear. The aim of the study was to identify latent comorbidity patterns and explore the comorbidity network structure that included 12 chronic conditions in adults diagnosed with depressive disorder. A cross-sectional study was conducted based on secondary data from the 2017 behavioral risk factor surveillance system (BRFSS) covering all 50 American states. A sample of 89,209 U.S. participants, 29,079 men and 60,063 women aged 18 years or older, was considered using exploratory graphical analysis (EGA), a statistical graphical model that includes algorithms for grouping and factoring variables in a multivariate system of network relationships. The EGA findings show that the network presents 3 latent comorbidity patterns, i.e., that comorbidities are grouped into 3 factors. The first group was composed of 7 comorbidities (obesity, cancer, high blood pressure, high blood cholesterol, arthritis, kidney disease, and diabetes). The second pattern of latent comorbidity included the diagnosis of asthma and respiratory diseases. The last factor grouped 3 conditions (heart attack, coronary heart disease, and stroke). Hypertension reported higher measures of network centrality. Associations between chronic conditions were reported; furthermore, they were grouped into 3 latent dimensions of comorbidity and reported network factor loadings. The implementation of care and treatment guidelines and protocols for patients with depressive symptomatology and multimorbidity is suggested.

Circulation of a digital community currency

Circulation is the characteristic feature of successful currency systems, from community currencies to cryptocurrencies to national currencies. In this paper, we propose a network analysis approach especially suited for studying circulation given a system’s digital transaction records. Sarafu is a digital community currency that was active in Kenya over a period that saw considerable economic disruption due to the COVID-19 pandemic. We represent its circulation as a network of monetary flow among the 40,000 Sarafu users. Network flow analysis reveals that circulation was highly modular, geographically localized, and occurring among users with diverse livelihoods. Across localized sub-populations, network cycle analysis supports the intuitive notion that circulation requires cycles. Moreover, the sub-networks underlying circulation are consistently degree disassortative and we find evidence of preferential attachment. Community-based institutions often take on the role of local hubs, and network centrality measures confirm the importance of early adopters and of women’s participation. This work demonstrates that networks of monetary flow enable the study of circulation within currency systems at a striking level of detail, and our findings can be used to inform the development of community currencies in marginalized areas.

Directed communication in games with directed graphs

AbstractWe introduce a novel concept of directed communication and a related connectedness in directed graphs, and apply this to model certain cooperation restrictions in cooperative games. In the literature on communication in directed networks or directed graphs, one can find different notions of connectedness, and different ways how directed communication restricts cooperation possibilities of players in a game. In this paper, we introduce a notion of connectedness in directed graphs that is based on directed paths. We assume that a coalition of players in a game can only cooperate if these players form a directed path in a directed communication graph. We define a restricted game following the same approach as Myerson for undirected communication situations, and consider the allocation rule that applies the Shapley value to this restricted game. We characterize this value by extended versions of the well-known component efficiency, fairness and balanced contributions axioms. Moreover, using the new notion of connectedness, we apply this allocation rule to define network centrality, efficiency and vulnerability measures for directed networks.

Computational prediction of protein folding rate using structural parameters and network centrality measures

Protein folding is a complex physicochemical process whereby a polymer of amino acids samples numerous conformations in its unfolded state before settling on an essentially unique native three-dimensional (3D) structure. To understand this process, several theoretical studies have used a set of 3D structures, identified different structural parameters, and analyzed their relationships using the natural logarithmic protein folding rate (ln(kf)). Unfortunately, these structural parameters are specific to a small set of proteins that are not capable of accurately predicting ln(kf) for both two-state (TS) and non-two-state (NTS) proteins. To overcome the limitations of the statistical approach, a few machine learning (ML)-based models have been proposed using limited training data. However, none of these methods can explain plausible folding mechanisms. In this study, we evaluated the predictive capabilities of ten different ML algorithms using eight different structural parameters and five different network centrality measures based on newly constructed datasets. In comparison to the other nine regressors, support vector machine was found to be the most appropriate for predicting ln(kf) with mean absolute differences of 1.856, 1.55, and 1.745 for the TS, NTS, and combined datasets, respectively. Furthermore, combining structural parameters and network centrality measures improves the prediction performance compared to individual parameters, indicating that multiple factors are involved in the folding process.

A new fast entropy‐based method to generate composite centrality measures in complex networks

SummaryDetermining the centrality of nodes in complex networks provides practical benefits in many areas such as detecting influencer nodes, viral marketing, and preventing the spread of rumors. On the other hand, there is no consensus for the definition of centrality. Therefore, different centrality measures such as degree, closeness, and betweenness have been developed to measure the centrality of a node. However, each centrality measure highlights the various characteristics of the nodes in the network from its own point of view. This causes each centrality measure to rank the nodes in a different order. In recent years, researchers have focused on approaches that combine multiple centrality measures. Thus, the perspectives of different centrality measures can be considered simultaneously. In this study, we have proposed a fast and efficient method using the analytic hierarchy process and entropy weighting to combine multiple centrality measures. We tested the proposed method with synthetic and real datasets and compared the results with those of state‐of‐the‐art methods. The experimental results showed the proposed method to be competitive with these advanced methods, whereas it performed much better than the other methods in terms of computational speed. This indicated that our proposed method could be applied to large and dynamic complex networks.

Twenty-Four-Hour Movement Behaviors, Fitness, and Adiposity in Preschoolers: A Network Analysis

The present study aimed to verify the associations between compliance with the 24-h movement behavior recommendations, fitness, and adiposity markers in preschoolers, considering the non-linear nature of these associations. The sample was comprised of 253 preschoolers. Preschoolers were assessed for anthropometric data and wore an accelerometer for seven consecutive days. Screen time and sleep duration were parent-reported in a face-to-face interview. The PREFIT test battery was used to assess physical fitness components (lower-body strength, cardiorespiratory fitness, and speed/agility). Descriptive statistics were used to describe the variables, and a network analysis was conducted to assess the emerging pattern of associations between the variables. Preschoolers’ greatest compliance with recommendations was observed for physical activity, while the lowest compliance was observed for the screen time recommendation. Among children aged three years, only 2.2% complied with all recommendations; only 1.0% of the four-year-olds and 1.3% of the five-year-olds complied with all recommendations. The results of the network analysis and centrality measures emphasized that cardiorespiratory fitness (CRF) and compliance with movement behavior recommendations were the most critical variables to address in preschoolers, reinforcing the importance of intervention programs focused on intense activities.

An efficient weighted network centrality approach for exploring mechanisms of action of the Ruellia herbal formula for treating rheumatoid arthritis

AimThis study outlines an efficient weighted network centrality measure approach and its application in network pharmacology for exploring mechanisms of action of the Ruellia prostrata (RP) and Ruellia bignoniiflora (RB) herbal formula for treating rheumatoid arthritis.MethodIn our proposed method we first calculated interconnectivity scores all the network targets then computed weighted centrality score for all targets to identify of major network targets based on centrality score. We apply our technology to network pharmacology by constructing herb-compound-putative target network; compound-putative targets-RA target network; and imbalance multi-level herb-compound-putative target-RA target-PPI network. We then identify the major targets in the network based on our centrality measure approach. Finally we validated the major identified network targets using the enrichment analysis and a molecular docking simulation.ResultThe results reveled our proposed weighted network centrality approach outperform classical centrality measure in identification of influential nodes in four real complex networks based on SI model simulation. Application of our approach to network pharmacology shows that 57 major targets of which 33 targets including 8 compositive compounds, 15 putative target and 10 therapeutic targets played an important role in the network and directly linked to rheumatoid arthritis. Enrichment analysis confirmed that putative targets were frequently involved in TNF, CCR5, IL-17 and G-protein coupled receptors signaling pathways which are critical in the progression of rheumatoid arthritis. The molecular docking simulation indicated four targets had significant binding affinity to major protein targets. Glyceryl diacetate-2-Oleate and Oleoyl chloride showed the best binding affinity to all targets proteins and were within Lipinski limits. ADMET prediction also confirm both compounds had no toxic effect on human hence potential lead drug compounds for treating rheumatoid arthritis.ConclusionThis study developed an efficient weighted network centrality approach as tool for identification of major network targets. Network pharmacology findings provides promising results that could lead us to design and discover of alternative drug compounds. Though our approach is a purely in silico method, clinical experiments are required to test and validate the hypotheses of our computational methods.

Centrality measures in fuzzy social networks

Centrality measures have been widely used to capture the properties of different nodes in a social network, particularly when the edges are fully deterministic. Various models have also been proposed to calculate nodes’ centrality in graphs where there might be some uncertainty in relation to the edges. Their common characteristic is that graph uncertainty is essentially embedded into the calculation of centrality to compute a single crisp value. However, as the degree of uncertainty may vary, centrality values may also vary. In this paper, making use of fuzzy set theory, we assume that a social network is modelled by a fuzzy graph and a fuzzy variable is used to describe the truth degree of an edge between two nodes. Based on this formulation, appropriate definitions are given to determine the truth degree of different centrality values for a node and thereby centrality as a fuzzy relation. Three well-known centrality measures, degree, h-index and k-shell, are extended to calculate the truth degree for the centrality of a node in a fuzzy graph. Experimental results demonstrate that the proposed centrality measures can determine the importance of nodes in a fuzzy graph more accurately than other fuzzy or deterministic centrality measures.

Global value chains and domestic innovation

This paper explores how changes in both position and participation in Global Value Chain (GVC) networks affect firm innovation. The analysis combines matched patent-firm data for Japan with measures of GVC network centrality and GVC participation using the OECD Inter-Country Input-Output (ICIO) Tables over the period from 1995 to 2011. We find that Japan's position in GVCs has shifted from being at the core of Asian value chains towards the periphery relative to other countries in the network, i.e., becoming less “central”. We use China's accession to the World Trade Organization as an instrumental variable for changes in Japanese centrality. Our analysis shows that increases in Japanese sectors' forward centrality – i.e. as a key supplier - tend to be positively associated with increasing firms' patent applications in these sectors and that firms in key hubs within GVCs, specifically as key suppliers, appear to benefit from knowledge spillovers from downstream markets.

On the use of graph centralities to compute generalized inverse of singular finite element operators: Applications to the analysis of floating substructures

SummaryThis article introduces a robust and affordable method to compute nullspace and generalized inverse of finite element operators involved in dual domain decomposition methods. The methodology relies on the operator partial factorization and on the analysis of a well chosen Schur complement. The sparse linear operator is interpreted as a network and graph centrality measures are used to select the condensation variables. Eigenvector, Katz and Page Rank centralities are evaluated. An extension to deal with symmetric indefinite systems arising from mixed finite elements is also presented. The approach is assessed on highly heterogeneous problems and one industrial application is presented: the numerical homogenization of solid propellant.

Social diversity reduces the complexity and cost of fostering fairness

Institutions and investors are constantly faced with the challenge of appropriately distributing endowments. No budget is limitless and optimising overall spending without sacrificing positive outcomes has been approached and resolved using several heuristics. To date, prior works have failed to consider how to encourage fairness in a population where social diversity is ubiquitous, and in which investors can only partially observe the population. Herein, by incorporating social diversity in the Ultimatum game through heterogeneous graphs, we investigate the effects of several interference mechanisms which assume incomplete information and flexible standards of fairness. We quantify the role of diversity and show how it reduces the need for information gathering, allowing us to relax a strict, costly interference process. Furthermore, we find that the influence of certain individuals, expressed by different network centrality measures, can be exploited to further reduce spending if minimal fairness requirements are lowered. Our results indicate that diversity changes and opens up novel mechanisms available to institutions wishing to promote fairness. Overall, our analysis provides novel insights to guide institutional policies in socially diverse complex systems.

Psychopathological networks in psychosis: Changes over time and clinical relevance. A long-term cohort study of first-episode psychosis

BackgroundFirst-episode psychosis is a critical period for early interventions to reduce the risk of poor outcomes and relapse as much as possible. However, uncertainties about the long-term outcomes of symptomatology remain to be ascertained. MethodsThe aim of the present study was to use network analysis to investigate first-episode and long-term stages of psychosis at three levels of analysis: micro, meso and macro. The sample was a cohort of 510 patients with first-episode psychoses from the SEGPEP study, who were reassessed at the long-term follow-up (n = 243). We used the Comprehensive Assessment of Symptoms and History for their assessments and lifetime outcome variables of clinical relevance. ResultsOur results showed a similar pattern of clustering between first episodes and long-term follow-up in seven psychopathological dimensions at the micro level, 3 and 4 dimensions at the meso level, and one at the macro level. They also revealed significant differences between first-episode and long-term network structure and centrality measures at the three levels, showing that disorganization symptoms have more influence in long-term stabilized patients. ConclusionsOur findings suggest a relative clustering invariance at all levels, with the presence of two domains of disorganization as the most notorious difference over time at micro level. The severity of disorganization at the follow-up was associated with a more severe course of the psychosis. Moreover, a relative stability in global strength of the interconnections was found, even though the network structure varied significantly in the long-term follow-up. The macro level was helpful in the integration of all dimensions into a common psychopathology factor, and in unveiling the strong relationships of psychopathological dimensions with lifetime outcomes, such as negative with poor functioning, disorganization with high antipsychotic dose-years, and delusions with poor adherence to treatment. These results add evidence to the hierarchical, dimensional and longitudinal structure of psychopathological symptoms and their clinical relevance in first-episode psychoses.