Large-scale Social Networks Research Articles

Stochastic bilevel interdiction for fake news control in online social networks

Social media platforms attempt to mitigate and control fake news, using interventions such as flagging posts or adjusting newsfeed algorithms, to protect vulnerable individuals. In this paper, we consider performing intervention actions on specific source nodes or user–user edges in social networks, under uncertain effectiveness of different intervention strategies. We model misinformation from malicious users to vulnerable communities using stochastic network interdiction formulations. Specifically, we minimize the expected number of reachable vulnerable users via stochastic maximum flow, and develop an alternative formulation for handling large-scale social networks based on their topological structures. We derive theoretical results for path-based networks and develop an approximate algorithm for single-edge removal on paths. We test instances of a social network with 23,505 nodes, based on the IMDb actors dataset, to demonstrate the scalability of the approach and its effectiveness. Via numerical studies, we find that characteristics of removed edges change when intervention effectiveness is stochastic. Our results suggest that intervention should target on (i) a smaller set of centrally located edges with nodes that represent communities where regulatory actions are more effective, and (ii) dispersed edges with nodes where intervention has a high chance of failure.

Consensus methods with Nash and Kalai-Smorodinsky bargaining game for large-scale group decision-making

With the significant advancements in communication technology, group decision-making (GDM) can now be implemented online, allowing a large number of decision-makers (DMs) to participate concurrently. However, current methods for large-scale group decision-making (LSGDM) are primarily suitable for 20 to 50 DMs, and their effectiveness in scenarios involving thousands or even tens of thousands of participants has yet to be fully validated. Furthermore, as the number of participants increases, the evaluation information becomes increasingly diverse and complex. At the same time, the social networks associated with the DMs typically become sparse, making information sharing and consensus building more challenging. In light of these challenges, we develop two new methods based on cooperative games to effectively address the challenges in super LSGDM. First, we propose a two-stage semi-supervised fuzzy C-means (FCM) clustering method with trust constraints, which aims to address the issue of sparsity in relationships within large-scale social networks. This method utilizes trust relationships as reliable resources and prior knowledge to guide and supervise the clustering process. On this basis, we discuss three scenarios from the perspective of cooperative games: (i) subgroup optimal consensus adjustments in non-cooperative situations, (ii) group optimal consensus adjustments in cooperative situations, and (iii) subgroup optimal consensus adjustments in cooperative situations. Subsequently, we view the consensus adjustment allocation as a cost cooperative game problem and propose two new LSGDM consensus methods based on Nash Bargaining (NB) and Kalai-Smorodinsky Bargaining (KSB). Finally, experiments on real datasets demonstrate the superiority and reliability of our proposed LSGDM methods.

Local core expanding-based label diffusion and local deep embedding for fast community detection algorithm in social networks

Community detection is a key task in social network analysis, as it reveals the underlying structure and function of the network. Various global and local techniques exist for uncovering community structures in social networks wherein diffusion-based algorithms are proposed as novel methods for local community detection, particularly suited for large-scale networks. The efficacy of diffusion processes and initial detection is paramount in the successful identification of community structures within social networks. This effectiveness hinges significantly on the meticulous selection of the label diffuser core, which serves as the foundation for propagating labels through the network, and the precise labeling of boundary nodes. Addressing the constraints of current community detection algorithms, notably their time complexity and efficiency, this paper proposes a novel local community detection algorithm that combines core expansion with label diffusion, and deep embedding techniques. In the proposed method, a new centrality measure is introduced for appropriate core selection to facilitate precise label diffusion in the initial phase. Subsequently, a deep embedding technique is employed for updating labels of boundary and core nodes using the GraphSage embedding method. Finally, a rapid merging step is executed to amalgamate initially proximate communities into finalized community structures in large-scale social networks. We evaluate our algorithm on 14 real-world and 4 synthetic networks and show that it outperforms existing methods in terms of NMI, F-measure, ARI, and modularity. According to numerical results, the proposed method shows approximately 1.04 %, 1.03 %, and 1.12 % improvement in F-measure, NMI, and ARI measures respectively, compared to the second-best method, LBLD, in the networks with ground-truth. In addition, our method is able to accurately identify communities in large-scale networks such as Orkut, YouTube, and LiveJournal, where it ranks among the top-performing methods. Our approach exhibits the best performance in terms of ARI compared to other algorithms under comparison.

Emotion contagion and stock market: the role of online investor network in China

ABSTRACT Social media serves as a crucial channel for investors to acquire asset-related information while also facilitating emotional contagion through social connections. However, the association between the emotional contagion on social networks and asset prices remains unclear. We examine the relationship between market returns and disseminated emotion resulting from emotion contagion in a social network of more than 2.5 million investors in a finance-specific forum. Investors’ emotions are detected from their short-text posts based on a novel classification framework that leverages the traditional dictionary-based method and the Bidirectional Encoder Representations from Transformers (BERT). The potential of investors in emotional contagion is approximated in terms of a global metric of Collective Influence measured daily from a large-scale social network. We find that disseminated emotion is negatively associated with next-day returns, highlighting the importance of the network effect in reshaping the process of investors’ emotional contagion and subsequently influencing asset prices. Our findings suggest that understanding and monitoring emotional contagion in social networks can aid in predicting market movements, and policymakers can leverage these insights to enhance market stability by addressing investors’ psychological and social aspects.

Node-Level Community Detection within Edge Exchangeable Models for Interaction Processes

Scientists are increasingly interested in discovering community structure from modern relational data arising on large-scale social networks. While many methods have been proposed for learning community structure, few account for the fact that these modern networks arise from processes of interactions in the population. We introduce block edge exchangeable models (BEEM) for the study of interaction networks with latent node-level community structure. The block vertex components model (B-VCM) is derived as a canonical example. Several theoretical and practical advantages over traditional vertex-centric approaches are highlighted. In particular, BEEMs allow for sparse degree structure and power-law degree distributions within communities. Our theoretical analysis bounds the misspecification rate of block assignments while supporting simulations show the properties of the network can be recovered. A computationally tractable Gibbs algorithm is derived. We demonstrate the proposed model using post-comment interaction data from Talklife, a large-scale online peer-to-peer support network, and contrast the learned communities from those using standard algorithms including degree-corrected stochastic block models, popularity-adjusted block models, and weighted stochastic block models. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work.

Bayesian Analysis of Exponential Random Graph Models Using Stochastic Gradient Markov Chain Monte Carlo

The exponential random graph model (ERGM) is a popular model for social networks, which is known to have an intractable likelihood function. Sampling from the posterior for such a model is a long-standing problem in statistical research. We analyze the performance of the stochastic gradient Langevin dynamics (SGLD) algorithm (also known as noisy Longevin Monte Carlo) in tackling this problem, where the stochastic gradient is calculated via running a short Markov chain (the so-called inner Markov chain in this paper) at each iteration. We show that if the model size grows with the network size slowly enough, then SGLD converges to the true posterior in 2-Wasserstein distance as the network size and iteration number become large regardless of the length of the inner Markov chain performed at each iteration. Our study provides a scalable algorithm for analyzing large-scale social networks with possibly high-dimensional ERGMs.

IMNE: Maximizing influence through deep learning-based node embedding in social network

Influence Maximization (IM) is a critical problem in social network analysis and marketing. It involves identifying a subset of nodes in a social network whose activation or influence can lead to the maximal spread of information, ideas, or behaviors within the network. Although many approaches have been developed in the literature to deal with this problem, most of these approaches are ineffective in dealing with large-scale social networks due to free parameters and computational complexity. Embeddings are used to learn low-dimensional representations of nodes in a social network. These embeddings capture the structural and semantic information of nodes and their relationships within the network. By training deep learning models on graph-structured data, node embeddings can capture complex patterns and dependencies in social networks, enabling more effective downstream tasks such as IM. Accordingly, this paper proposes an efficient algorithm to address the IM problem in social networks using deep learning-based Node Embedding (IMNE), which includes shell decomposition, graph/node embedding, and search space reduction as well as the use of local structural features. Our approach combines the power of deep learning for representation learning with the rich structural information present in social networks to address the challenge of IM in complex and dynamic social networks. IMNE uses the Independent Cascade (IC) information diffusion model to determine the labels needed to train the model by calculating the influence of nodes. Experimental results on several real-world networks considering different performance metrics show that IMNE performs better compared to existing baseline and state-of-the-art methods.

FairSNA: Algorithmic Fairness in Social Network Analysis

In recent years, designing fairness-aware methods has received much attention in various domains, including machine learning, natural language processing, and information retrieval. However, in social network analysis (SNA), designing fairness-aware methods for various research problems by considering structural bias and inequalities of large-scale social networks has not received much attention. In this work, we highlight how the structural bias of social networks impacts the fairness of different SNA methods. We further discuss fairness aspects that should be considered while proposing network structure-based solutions for different SNA problems, such as link prediction, influence maximization, centrality ranking, and community detection. This survey-cum-vision clearly highlights that very few works have considered fairness and bias while proposing solutions; even these works are mainly focused on some research topics, such as link prediction, influence maximization, and PageRank. However, fairness has not yet been addressed for other research topics, such as influence blocking and community detection. We review the state of the art for different research topics in SNA, including the considered fairness constraints, their limitations, and our vision. This survey also covers evaluation metrics, available datasets and synthetic network generating models used in such studies. Finally, we highlight various open research directions that require researchers’ attention to bridge the gap between fairness and SNA.

A Graph Sampling-Based Model for Influence Maximization in Large-Scale Social Networks

Social networks have attracted a great deal of attention and have, in fact, changed the way we produce, consume, and diffuse information. This change gave rise to the notion of social influence and today we talk about influential nodes. The process of detecting influential nodes in social networks aims to find entities that propagate information to a large portion of the network users. This process is often known as the influence maximization (IM) problem. Due to the explosive growth of social networks’ data, their structure is more complex and we talk about “big graph data.” Moreover, modern networks are dynamic and their topology or/and information is likely to change over time. Detecting influential nodes in such networks is a challenging task. Several methods have been developed in this context. However, they concentrate on static networks and there is little work on large-scale social networks. We propose in this article a new model for IM called MapReduce-based dynamic selection of influential nodes (MR-DSINs) that has the ability to cope with the huge size of real social networks. In fact, our approach is based on a graph sampling step in order to reduce the network’s size. Given that reduced version, MR-DSIN is able to select dynamically influential nodes. Our proposal has the advantage of considering the dynamics of information that can be modeled by users’ social actions (e.g.“, share”, “comment”, “retweet”). Experimental results on real-world social networks and computer-generated artificial graphs demonstrate that MR-DSIN is efficient for identifying influential nodes, compared with three known proposals. We prove that our model is able to detect in the reduced graph an influence as important as in the original one.

Disease avoidance threatens social cohesion in a large-scale social networking experiment

People tend to limit social contacts during times of increased health risks, leading to disruption of social networks thus changing the course of epidemics. To what extent, however, do people show such avoidance reactions? To test the predictions and assumptions of an agent-based model on the feedback loop between avoidance behavior, social networks, and disease spread, we conducted a large-scale (2,879 participants) incentivized experiment. The experiment rewards maintaining social relations and structures, and penalizes acquiring infections. We find that disease avoidance dominates networking decisions, despite relatively low penalties for infections; and that participants use more sophisticated strategies than expected (e.g., avoiding susceptible others with infectious neighbors), while they forget to maintain a beneficial network structure. Consequently, we observe low infection numbers, but also deterioration of network positions. These results imply that the focus on a more obvious signal (i.e., infection) may lead to unwanted side effects (i.e., loss of social cohesion).

Emergence of a mutual-growth mechanism in networks evolved by social preference based on indirect utility

Preferential attachment is an important mechanism in the structural evolution of complex networks. However, though resources on a network propagate and have an effect beyond a direct relationship, growth by preferential attachment based on indirectly propagated resources has not been systematically investigated. Here, we propose a mathematical model of an evolving network in which preference is proportional to a utility function reflecting direct utility from directly connected nodes and indirect utility from indirectly connected nodes beyond the directly connected nodes. Our analysis showed that preferential attachment involving indirect utility forms a converged and hierarchical structure, thereby significantly increasing the indirect utility across the entire network. Further, we found that the structures are formed by mutual growth between adjacent nodes, which promotes a scaling exponent of 1.5 between the number of indirect and direct links. Lastly, by examining several real networks, we found evidence of mutual growth, especially in social networks. Our findings demonstrate a growth mechanism emerging in evolving networks with preference for indirect utility, and provide a foundation for systematically investigating the role of preference for indirect utility in the structural and functional evolution of large-scale social networks.

An empirical study of the impact of biological information dissemination in social media on public science literacy

Abstract In this paper, we first establish a locally converged bioinformatics dataset based on gradient sampling and design an optimal data mining control model to improve the accuracy of bioinformatics big data feature mining. The performance of the Compressive Tracking algorithm and Online Bosting algorithm is compared with the mining error as a test index. At the same time, we propose a social media information dissemination algorithm applicable to large-scale social network datasets, taking the degree value of each node as the node’s full influence and comparing and analyzing the dissemination influence of BP-IM, RAND and MC-CELF algorithms. Finally, taking public health big data as the research object, the least squares regression method was used to analyze the influence of the amount of public attention to bioinformatics scientific knowledge on their scientific literacy in different media. The results showed that there was a significant positive correlation between scientific literacy and willingness to engage in science participation behavior on social media when the amount of public attention to scientific information was β =0225, p <0.01. When more people are interested in scientific knowledge of bioinformatics on social media, their scientific literacy will improve.

Social relation-driven consensus reaching in large-scale group decision-making using semi-supervised classification

The fundamental goal of group decision making (GDM) is to improve consensus amongst experts and reduce individual conflicts of interest in the process of alternative selection. By analysing the social network relationships of decision-makers (DMs), such as trust and preference similarity relationships, more effective consensus-reaching mechanisms can be designed. Previous studies developed supervised classification algorithms for DM groups of more than 1000 participants on social networks. However, in large-scale social networks, the social connection relations of DMs cannot be completely investigated as human resources and costs during the decision-making process impose limitations. These DMs cannot be effectively trained to build a classification model, which is referred to as unlabelled DMs and results in a large amount of information being ignored during the model training process, leading to an unfair process that can deepen the conflict amongst experts. To address the information loss problem, this study proposes a semi-supervised learning model for social networks with incomplete trust relations, aiming to effectively absorb the preference information of unlabelled DMs into the classification model, thereby improving the effectiveness of classification management. Specifically, the cost-sensitive semi-supervised support vector machine (SVM) introduced in the inner product space is subsequently used by the DMs' classification model to accurately divide unlabelled DMs. A minimum cost adjustment consensus model is then constructed based on the DM subgroups. Numerical experiments on urban renewal were conducted to verify the effectiveness of the proposed method. The empirical and simulation results demonstrated that the proposed method can decrease total consensus costs for social-network GDM with missing trust-relationship information.

Efficient community-based influence maximization in large-scale social networks

Efficient community-based influence maximization in large-scale social networks

Resampling reduces bias amplification in experimental social networks.

Large-scale social networks are thought to contribute to polarization by amplifying people's biases. However, the complexity of these technologies makes it difficult to identify the mechanisms responsible and evaluate mitigation strategies. Here we show under controlled laboratory conditions that transmission through social networks amplifies motivational biases on a simple artificial decision-making task. Participants in a large behavioural experiment showed increased rates of biased decision-making when part of a social network relative to asocial participants in 40 independently evolving populations. Drawing on ideas from Bayesian statistics, we identify a simple adjustment to content-selection algorithms that is predicted to mitigate bias amplification by generating samples of perspectives from within an individual's network that are more representative of the wider population. In two large experiments, this strategy was effective at reducing bias amplification while maintaining the benefits of information sharing. Simulations show that this algorithm can also be effective in more complex networks.

Simplifying social networks via triangle-based cohesive subgraphs

One main challenge for simplifying node-link diagrams of large-scale social networks lies in that simplified graphs generally contain dense subgroups or cohesive subgraphs. Graph triangles quantify the solid and stable relationships that maintain cohesive subgraphs. Understanding the mechanism of triangles within cohesive subgraphs contributes to illuminating patterns of connections within social networks. However, prior works can hardly handle and visualize triangles in cohesive subgraphs. In this paper, we propose a triangle-based graph simplification approach that can filter and visualize cohesive subgraphs by leveraging a triangle-connectivity called k-truss and a force-directed algorithm. We design and implement TriGraph, a web-based visual interface that provides detailed information for exploring and analyzing social networks. Quantitative comparisons with existing methods, two case studies on real-world datasets, and feedback from domain experts demonstrate the effectiveness of TriGraph.

Fair Influence Maximization in Large-scale Social Networks Based on Attribute-aware Reverse Influence Sampling

Influence maximization is the problem of finding a set of seed nodes in the network that maximizes the influence spread, which has become an important topic in social network analysis. Conventional influence maximization algorithms cause “unfair" influence spread among different groups in the population, which could lead to severe bias in public opinion dissemination and viral marketing. To address this issue, we formulate the fair influence maximization problem concerning the trade-off between influence maximization and group fairness. For the purpose of solving the fair influence maximization problem in large-scale social networks efficiently, we propose a novel attribute-based reverse influence sampling (ABRIS) framework. This framework intends to estimate influence in specific groups with guarantee through an attribute-based hypergraph so that we can select seed nodes strategically. Therefore, under the ABRIS framework, we design two different node selection algorithms, ABRIS-G and ABRIS-T. ABRIS-G selects nodes in a greedy scheduling way. ABRIS-T adopts a two-phase node selection method. These algorithms run efficiently and achieve a good trade-off between influence maximization and group fairness. Extensive experiments on six real-world social networks show that our algorithms significantly outperform the state-of-the-art approaches. This article appears in the AI & Society track.

Real-time updating of dynamic social networks for COVID-19 vaccination strategies

Vaccination strategy is crucial in fighting the COVID-19 pandemic. Since the supply is still limited in many countries, contact network-based interventions can be most powerful to set an efficient strategy by identifying high-risk individuals or communities. However, due to the high dimension, only partial and noisy network information can be available in practice, especially for dynamic systems where contact networks are highly time-variant. Furthermore, the numerous mutations of SARS-CoV-2 have a significant impact on the infectious probability, requiring real-time network updating algorithms. In this study, we propose a sequential network updating approach based on data assimilation techniques to combine different sources of temporal information. We then prioritise the individuals with high-degree or high-centrality, obtained from assimilated networks, for vaccination. The assimilation-based approach is compared with the standard method (based on partially observed networks) and a random selection strategy in terms of vaccination effectiveness in a SIR model. The numerical comparison is first carried out using real-world face-to-face dynamic networks collected in a high school, followed by sequential multi-layer networks generated relying on the Barabasi-Albert model emulating large-scale social networks with several communities.

Determination of influential nodes based on the Communities’ structure to maximize influence in social networks

With the increasing development of social networks, they have turned into important research platforms. Influence maximization is one of the most important research issues in the field of social networks. This problem detects influential k-node with the greatest influence spread. The influence maximization faces two important challenges, time efficiency and optimal selection of seed nodes. In order to solve such challenges, we propose an algorithm based on optimal pruning and scoring adjustment, which is called IMBC for short. The IMBC (Influence Maximization Based on Community structure) algorithm uses optimal pruning and a minimum of dominating nodes to improve time efficiency. In addition, for optimal selection of seed nodes, the IMBC algorithm modulates the scores of nodes with a high Rich-Club coefficient. In order to select influential nodes, we first select an optimal set using the minimum dominating nodes and node scores, with the aim of optimal pruning in influence spread calculations. Because large-scale social networks have many nodes, optimal pruning reduces computational overhead. Then, the seed nodes are selected based on the scoring adjustment. Scoring adjustment is done to avoid the Rich Club phenomenon because avoiding this phenomenon causes a large amount of diffusion in social networks. The experimental results show that the proposed algorithm performs better than the algorithms presented in recent years in influence spread and runtime. Therefore, the IMBC algorithm is a balance between quality and efficiency. Also, in the PGP dataset results, the PHG algorithm with as much as a 5.08% increase in influence spread, and the runtime has decreased by 97%.

Identifying influential nodes in large scale social networks using global and local structural information

Identifying influential nodes in large scale social networks using global and local structural information