Degree correlations in signed social networks
Introduction
Over the last few years, an increasing interest in the study of social networks has prompted physicists, mathematicians and computer scientists to join sociologists in their endeavors to develop network models concerned with the antecedents, structure, and evolution of social interaction [1], [2], [3]. Recent studies have indicated that social networks across many empirical domains display the typical signature of complex networks, namely the long-tailed distribution of the degrees of nodes [3]. In addition to this, an attempt has been made to uncover the distinctive structural features and empirical regularities that distinguish social networks from other types of complex networks. While in most real networks degrees of neighboring nodes tend to be anticorrelated, research has suggested that social networks tend to be characterized by the opposite correlation pattern [4], [5]. The tendency of nodes with similar degree to connect with each other is often referred to as “assortative mixing by degree”, and has been observed in a number of social networks, including very large-scale online social networks such as Facebook and Twitter [6].
A variety of models have been proposed by physicists, sociologists and computer scientists to explain these distinctive properties of social networks. For instance, assortative mixing has been related to the underlying community structure of social networks [5]. More recently, assortative mixing has been explained in terms of transitivity [7], homophily [8], and unsubstitutability of individuals and resources [9]. Research has also uncovered distinctive interaction patterns within social signed networks in which relationships can have a positive (e.g., trust and friendship) or negative (e.g., distrust and enmity) connotation [10]. In particular, the theory of “structural balance” has long suggested that, in undirected signed social networks, individuals embedded within closed triads tend to minimize cognitive tension: an individual tends to befriend a friend’s friend, distrust a friend’s enemy, befriend an enemy’s enemy, and distrust an enemy’s friend [11], [12].
Here we focus our attention on the emergence of degree correlations in signed networks, and how these correlations can be used to predict the sign of links in cases where it is not known or cannot be assessed directly. Indeed, despite the ubiquity and salience of negative relationships in a wide range of social systems, the detection of mixing patterns by degree has been confined primarily to the domain of unsigned networks or simply networks in which nodes were assumed to be connected through positive links (e.g., scientific collaboration networks and interlocking directorate networks [5], [9]). However, negative networks may exhibit correlation patterns that differ from those detected in positive networks [13]. Do individuals who distrust many others tend to distrust each other, or do they channel their negative feelings toward other individuals who distrust only very few others? To address this problem, here we propose a class of simple models that help uncover the relation between the sign of links and the type of degree correlations characterizing a network.
The outline of the paper is as follows. In Section 2, we introduce two signed online social networks, and examine the degree distributions and correlations of the positive and negative subnetworks extracted from the data. In Section 3, we propose a generative model of signed networks that polarize into two mutually exclusive groups of nodes. Section 3.1 focuses on the case of random networks with binomial degree distributions, whereas Section 3.2 deals with more realistic cases of networks with power-law degree distributions. Finally, in Section 4 we extend our modeling framework to networks in which nodes can split into three (or more) hostile groups. In Section 5, we summarize our findings and discuss their implications for research on signed complex networks.
Section snippets
The data
We analyze two online social networks. The first is the network formed by the users of Epinions (www.epinions.com), a website for user-generated reviews of various products. Registered users of Epinions can declare their trust or distrust toward one another, based on the comments they post. The second social network is formed by the users of Slashdot (www.slashdot.org), a website devoted to the discussion of technology-related news, and in which the Slashdot Zoo feature enables users to tag one
Signed networks with degree correlations that depend on the sign of the links
We begin by focusing on signed random networks with binomial degree distributions, in which nodes can be split into two mutually exclusive groups. Subsequently, we shall refine our analysis by investigating the case of assortative and disassortative signed networks with power-law degree distributions.
Extending the model: the case of three (or more) groups
Following the theoretical avenue that led Davis [31] to generalize the formalization of the theory of structural balance, we extend our model with network polarization to also account for the case in which nodes can be allocated to three or more mutually exclusive groups. As observed by Davis [31], individuals often split into more than two mutually hostile groups. To take this into account, Davis provided a generalization of the structure theorem [11], [21] by uncovering the necessary and
Conclusions
Our study was prompted by the empirical analysis of two signed social networks and by the observation that their mixing patterns by degree vary depending on the sign of the link. In particular, our findings indicated that negative subnetworks are characterized by disassortative patterns, in sharp contrast with their corresponding unsigned networks and the positive subnetworks. The emergence of opposite trends of mixing patterns seems to be at variance with the widely accepted belief that social
Acknowledgments
The authors acknowledge useful discussions with V. Loreto and V.D.P. Servedio.
References (31)
- et al.
Social structure of Facebook networks
Physica A
(2012) - et al.
Structure and time evolution of an Internet dating community
Social Networks
(2004) - et al.
Measuring social dynamics in a massive multiplayer online game
Social Networks
(2010) - et al.
A partitioning approach to structural balance
Social Networks
(1996) - et al.
Partitioning signed social networks
Social Networks
(2009) - et al.
On random graphs
Publ. Math. Debrecen
(1959) - et al.
Collective dynamics of small-world networks
Nature
(1998) Emergence of scaling in random networks
Science
(1999)Assortative mixing in networks
Phys. Rev. Lett.
(2002)- et al.
Why social networks are different from other types of networks
Phys. Rev. E
(2003)
The anatomy of the Facebook social graph, Tech. Rep.
Clustering drives assortativity and community structure in ensembles of networks
Phys. Rev. E
Multirelational organization of large-scale social networks in an online world
Proc. Natl. Acad. Sci. USA
Structural balance: a generalization of Heider’s theory
Psychol. Rev.
Attitudes and cognitive organization
J. Psychol.
Cited by (19)
Threshold cascade dynamics on signed random networks
2023, Chaos, Solitons and FractalsSocial effects of topic propagation on Weibo
2022, Journal of Management Science and EngineeringCitation Excerpt :For example, Klein et al. (2015) analyzed node activities to identify the critical nodes within online social networks and found that the personal activities of critical nodes are significantly correlated with their centrality within the online structure. Ciotti et al. (2015) examined degree correlations in two online social networks, wherein users were connected through different types of links. Leskovee (2007) and Liben-Nowell and Kleinberg (2008) focused on information propagation pathways and hierarchies in online social networks, while Fang et al. (2014) studied structural attributes, development patterns, groups and their interactions, and information dissemination in online social networks.
A physical pathway to understand individual's labeling behavior in signed social networks
2018, Physics Letters, Section A: General, Atomic and Solid State PhysicsCitation Excerpt :Those social networks involving both positive and negative links can be represented in terms of signed social networks [5–8].
How to estimate the signs' configuration in the directed signed social networks?
2017, Physics Letters, Section A: General, Atomic and Solid State PhysicsCitation Excerpt :For example, users can tag directed relations to others indicating trust or distrust in the trust network of Epinions, and users can designate others as “friends” or “foes” in the social network of the technology blog Slashdot [2]. Those social networks can be represented in terms of signed social networks [3–5], where a sign of link is defined as “+1” or “−1” depending on whether it expresses a positive or negative attitude from the generator of the link to the recipient [6]. The second issue about the role of the real signs' configuration in the dynamics of and on signed social networks has also been studied in the last decades [11–18].
Multi-objective particle swarm optimisation of complex product change plan considering service performance
2023, CAAI Transactions on Intelligence Technology