Sorting out centrality: An analysis of the performance of four centrality models in real and simulated networks

doi:10.1016/0378-8733(88)90014-7

Social Networks

Volume 10, Issue 3, September 1988, Pages 233-253

https://doi.org/10.1016/0378-8733(88)90014-7 Get rights and content

Abstract

Although the concept of centrality has been well developed in the social networks literature, its empirical development has lagged somewhat. This paper moves a step in that direction by assessing the performance of four centrality models under a variety of known and controlled situations. It begins by examining the assumptions underlying each model, as well as its behavior in a community influence network. It then assesses the robustness and sensitivity of each model under conditions of random and systematic variation introduced into this network.

References (9)

J.M. Bolland
Perceived leadership stability and the structure of urban agenda-setting networks
Social Networks
(1985)
F. Harary et al.
Structural Models
(1965)
A. Bavelas
A mathematical model for group structures
Applied Anthropology
(1948)
P. Bonacich
Factoring and weighting approaches to status scores and clique identification
Journal of Mathematical Sociology
(1971)
P. Bonacich
Power and centrality: a family of measures
American Journal of Sociology
(1987)

There are more references available in the full text version of this article.

Cited by (94)

Urbanization does not endanger food security: Evidence from China's Loess Plateau
2023, Science of the Total Environment
The expansion of construction land due to urbanization is the most rapid land use change in contemporary human history and has always occupied high-quality cropland, posing a severe threat to cropland and food security, it's essential to clarify the impact of urbanization on cropland and food security. This study proposed a research framework based on the regulating role of human activities, used quantifiable complex network analysis to uncover the vital role of urbanization in the evolution of land systems, and combined trajectory analysis of crop yield change to explore the impact of different urbanization modes (urban, town, and township mode) on food security through a continuous observation on 4259 township-level administrative regions of the Loess Plateau from 1990 to 2020. The findings proved that urbanization occupied the greatest land use area of cropland, and the town mode occupied 58.62 % of all urbanization modes encroaching on cropland, which has become a new pattern for advancing the urbanization development in the Loess Plateau. Construction land is more likely to be transferred in than out in the land use transfer network, while other land use types converted to construction land will be difficult to reverse. The Chinese government has implemented pragmatic policies, improved agricultural production techniques, and promoted agricultural intensification, resulting in a considerable increase in crop productivity and crop yield and the achievement of basic crop yield self-sufficiency of the Loess Plateau, so urbanization would not endanger food security. This study not only provides a more systematic research framework for related studies but also provides a theoretical basis for securing food security in other rapidly urbanizing regions of the world.
A parameterizable influence spread-based centrality measure for influential users detection in social networks
2022, Knowledge-Based Systems
In social network analysis, centrality refers to the relevance of actors or nodes within a social network represented as a graph. Traditional centrality measures are based on topological aspects of the network or the information flow circulating through it. Since 2010, new centrality measures have been proposed based on the influence spread capacity of the actors. One of the purposes of centrality measures is to obtain well-differentiated rankings among the actors. In order to achieve the above, recent experiments suggest that the actors must rely on some neighbors to increase their initial influence spread capacity. This article presents the General Influence Spread Rank (GISR), a generalized and parameterizable centrality measure based on two well-known influence spread models: the Independent Cascade model and the Linear Threshold Model. This measure allows adjusting the depth levels, probability, and directionality of the initial activation of neighbors, to study how these parameters intervene in the actors’ centrality. The experiments validate the importance of the depth level of neighborhoods and reveal that other connectivity aspects are also relevant to the centrality problem. Further, the depth level seems to be the most relevant variable in the influence spread for denser networks. In contrast, for sparser networks, the probability of selecting neighbors seems even more significant than the depth level.
Unearthing vulnerability of supply provision in logistics networks to the black swan events: Applications of entropy theory and network analysis
2021, Reliability Engineering and System Safety
Over the past two decades, the fragility of modern Logistics Networks (LNs) has been exposed by unexpected events with extreme impacts outside the realm of regular expectations, which are known as the black swan events. As a response to the growing dysfunctions of the global LNs due to the detrimental effects of the black swan events, this article develops a quantitative vulnerability assessment method by paying simultaneous attention to various structural properties of these networks. It adopts three prototypical examples of centrality indices known as betweenness, closeness, and eigenvector centrality. This work also develops a vulnerability index from the joint entropy of centrality values, which can be interpreted as a certain extent of risk to disruptions in supply provision in LNs. The proposed entropy-based vulnerability index enables a far more accurate analysis of vulnerability by measuring the degree of homogeneity and heterogeneity of centrality values. The suggested index is considered as an objective function that must be minimized to reduce the deleterious effects of the black swan events. To illustrate the proposed vulnerability assessment method, this paper analyzes a real-world global maritime network, concluding that the juxtaposition of centrality measures provides a richer insight into the vulnerability analysis of LNs. The results also favor the fact that increasing the homogeneity of a network through decentralization yields lower vulnerability and builds extra robustness into the networks.
Spare parts supply chain network modeling based on a novel scale-free network and replenishment path optimization with Q learning
2021, Computers and Industrial Engineering
The efficiency of inventory management determines the customers’ buying experience, so a supply chain network with a shorter replenishment time is needed. The supply chain network is hoped to be robust to the stock-out of some distributors in the network under random customer demands. At the same time, replenishment path optimization method with the objective of minimizing the replenishment time is required. After reviewing previous work done in the field of supply network topology, scale-free network is proven to be efficient when it was used to model supply network. In addition, multi-agent based collaborative replenishment model is smarter. But, there is rare research on multi-agent based collaborative replenishment in the supply chain modelled by scale-free network. In this study, we proposed a spare parts supply chain network model based on a novel scale-free network. In this network growth process, the connection probability function of connecting new distributor to the existing distributors in the network, is constructed considering the connection number (for an existing distributor, its connection number means the number of other distributors which have collaborative relationship with it) and inventory capacity of the existing distributors and the transit time between new distributor and existing distributors. The connection probability function is built from the standpoints of both new distributor and the existing distributors. Furthermore, different selection policies are discussed in the network growth process to improve the efficiency. Unlike other replenishment path optimization methods, Q learning takes the advantage of interacting with the environment to make a dynamic decision. So, Q learning is selected to optimize the replenishment path in supply chain network. In the experiment, network static and dynamic performance is analyzed using the indicators: degree distribution, clustering coefficient, centrality and response time. Experimental results showed that the replenishment time of supply chain network which are optimized by Q learning is reduced by approximately 40%. So, the shorter replenishment time of the supply chain network is verified.
Assessing the performance of the bootstrap in simulated assemblage networks
2021, Social Networks
Citation Excerpt :
Analysts have long been interested in aspects of variability in network actor centrality measures; here we briefly note some of that work. Bolland (1988) conducted an early study of centrality in simulated networks produced by randomly changing elements of the 0/1 adjacency matrix of an empirical network. Among other analyses, he compared mean correlations between centrality scores in the simulated and original empirical networks under different levels of randomness (up to 20 % of network elements being changed) for different centrality measures.
Archaeologists are increasingly interested in networks constructed from site assemblage data, in which weighted network ties reflect sites’ assemblage similarity. Equivalent networks would arise in other scientific fields where actors’ similarity is assessed by comparing distributions of observed counts, so the assemblages studied here can represent other kinds of distributions in other domains. One concern with such work is that sampling variability in the assemblage network and, in turn, sampling variability in measures calculated from the network must be recognized in any comprehensive analysis. In this study, we investigated the use of the bootstrap as a means of estimating sampling variability in measures of assemblage networks. We evaluated the performance of the bootstrap in simulated assemblage networks, using a probability structure based on the actual distribution of sherds of ceramic wares in a region with 25 archaeological sites. Results indicated that the bootstrap was successful in estimating the true sampling variability of eigenvector centrality for the 25 sites. This held both for centrality scores and for centrality ranks, as well as the ratio of first to second eigenvalues of the network (similarity) matrix. Findings encourage the use of the bootstrap as a tool in analyses of network data derived from counts.
Comparison of path-based centrality measures in protein-protein interaction networks revealed proteins with phenotypic relevance during adaptation to changing nitrogen environments
2021, Journal of Proteomics
Citation Excerpt :
These separated hubs can occur as a result of biological relevance (e.g. multiple self-sufficient protein-protein interaction complexes) or missing data. In addition, because BC is sensitive towards local topology, BC is more affected by missing data or random errors in edge-distribution than degree-based centrality measures [48]. On the other hand, because of its sensitivity, BC is most suited to observe true changes in network connectivity and node importance.
Plants must rapidly adapt to changes in nutrient conditions. Especially adaptations to changing nitrogen environments are very complex involving also major adjustments on the protein level. Here, we used a size-exclusion chromatography-coupled to mass spectrometry approach to study the dynamics of protein-protein interactions induced by transition from full nutrition to nitrogen starvation. Comparison of interaction networks established for each nutrient condition revealed a large overlap of proteins which were part of the protein-protein interaction network, but that same set of proteins underwent different interactions at each treatment. Network topology parameter betweenness centrality (BC) was found to best reflect the relevance of individual proteins in the information flow within each network. Changes in BC for individual proteins may therefore indicate their involvement in the cellular adjustments to the new condition. Based on this analysis, a set of proteins was identified showing high nitrogen-dependent changes in their BC values: The receptor kinase AT5G49770, co-receptor QSK1, and proton-ATPase AHA2. Mutants of those proteins showed a nitrate-dependent root growth phenotype. Individual interactions within the reconstructed network were tested using FRET-FLIM technology. Taken together, we present a systematic strategy comparing dynamic changes in protein-protein interaction networks based on their network parameters to identify regulatory nodes.
Protein-protein interactions are known to be important in cellular signaling events, but the dynamic changes in interaction networks induced by external stimuli are still rarely studied. We systematically analyzed how changes in the nutrient environment induced a rewiring of protein-protein interactions in roots. We observed small changes in overall protein abundances, but instead a rewiring of pairwise protein-protein interactions. Betweenness centrality was found to be the optimal network topology parameter to identify protein candidates with high relevance to the information flow in the (dynamic) network. Predicted interactions of those relevant nodes were confirmed in FLIM/FRET experiments and in phenotypic analysis. The network approach described here may be a useful application in dynamic network analysis more generally.

View all citing articles on Scopus

^☆: This is a revision of a paper originally presented at the 1986 meeting of the Sunbelt Social Network Conference. It benefitted from comments by Phil Bonacich, Wayne Baker, Kathleen Bolland, and anonymous reviewers.

^∗∗: Institute for Social Science Research, University of Alabama, P.O. Box 587, Tuscaloosa, AL 35487, U.S.A.

View full text

Sorting out centrality: An analysis of the performance of four centrality models in real and simulated networks☆

Abstract

Social Networks

A mathematical model for group structures

Applied Anthropology

Factoring and weighting approaches to status scores and clique identification

Journal of Mathematical Sociology

Power and centrality: a family of measures

American Journal of Sociology