Diffusion in Social Networks: A Multiagent Perspective

Jiang, Yichuan; Jiang, Jiuchuan

doi:10.1109/tsmc.2014.2339198

Cited by 105 publications

(9 citation statements)

References 101 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, the equality j∂f q /∂ω = ∂f q /∂γ holds. 4 Then, the above formulation reduces to dγ /dω = −j = 0, which reveals that γ does not exhibit an extremum at a regular point. To next investigate the possibility that γ exhibits an extremum at a singularity point, one can study ∂f q /∂ω = 0, which is necessary for the existence of an extremum in γ .…”

Section: Remark 25 (Numbering Of the Laplacian Eigenvalues)mentioning

confidence: 99%

“…The question of how dynamics, particles and systems interact with one another is a long-standing research topic, from traffic flow, swarms, animal groups to human behaviour [1][2][3][4]. A microscopic agent-based approach to study these interactions has led to a large body of literature in the context of 'multiagent systems' (MAS), especially in physics and engineering [5][6][7][8][9].…”

Section: Introductionmentioning

confidence: 99%

“…The continuity properties of the eigenvalues of a system with respect to its parameters indicate that a transition from γstability to γ -instability can only occur if the spectral abscissa crosses the shifted imaginary axis −γ + jω, see[39] 4. Let f q = R fq + jI fq and compute ∂f q /∂γ = ∂R fq /∂γ + j∂I fq /∂γ and ∂f q /∂ω = ∂R fq /∂ω + j∂I fq /∂ω.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Fast consensus in a large-scale multi-agent system with directed graphs using time-delayed measurements

Ramírez

Mondié

Garrido

2019

Phil. Trans. R. Soc. A.

View full text Add to dashboard Cite

This article is on fast-consensus reaching in a class of multi-agent systems (MAS). We present an analytical approach to tune controllers for the agents based on the premise that delayed measurements in the controller can be preferable to standard controllers relying only on current measurements. Controller tuning in this setting is however challenging due to the presence of delays. To tackle this problem, we propose an analytic geometry approach. The key contribution is that the tuning can be implemented for complex eigenvalues of the arising graph Laplacian of the network, complementing the current state of the art, which is limited to real eigenvalues. Results, therefore, extend our knowledge beyond symmetric graphs and enable the study of the MAS under directed graphs. This article is part of the theme issue ‘Nonlinear dynamics of delay systems’.

show abstract

Section: Remark 25 (Numbering Of the Laplacian Eigenvalues)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Fast consensus in a large-scale multi-agent system with directed graphs using time-delayed measurements

Ramírez

Mondié

Garrido

2019

Phil. Trans. R. Soc. A.

View full text Add to dashboard Cite

show abstract

“…There is a close relationship between multi-agent interaction and the diffusion of information [36], [37]. The main body of propagation can be modeled using interactive agents, the medium can be modeled as an interactive environment in the multi-agent system, the contents of the propagation can be modeled as a negotiation object in the multi-agent system, and the propagation model can be modeled as the interaction protocol and decision-making mechanism in the multi-agent system.…”

Section: Agent-based Information Diffusionmentioning

confidence: 99%

Structured Multi-Agent-Based Model for Bankruptcy Contagion With Cash Flow

Zhang

Xia

2020

IEEE Access

View full text Add to dashboard Cite

Numerous multi-agent models have been proposed for various economic phenomena, especially for the bankruptcy contagion phenomenon, which is a seriously destructive occurrence that begins with the bankruptcy of a small group of individuals and spreads to a large scale like an infectious disease. However, almost none of these bankruptcy models can be applied in plural environments or consider the difficulty of implementation. Furthermore, cash flow between firms, which is a highly influential factor in bankruptcy contagion situations, is considered in few such models. To address these shortcomings, in this paper, a new relationship, called the cash flow relationship, is first presented based on several relationships related to cash flows. A graphical structure called a cash flow graph is then presented to record cash flow relations between financial institutions and provide a discussion of the nature of the cash flow is then presented. Next, a multi-agent bankruptcy contagion model based on the cash flow graph is introduced. Finally, inferences are drawn from the proposed model and experiments conducted to explore the bankruptcy contagion phenomenon and confirm these inferences. This proposed model can be applied in multiple environments related to cash flows to successfully address the limitations of the existing multi-agent models for bankruptcy contagion. INDEX TERMS Agent decision-making; bankruptcy diffusion; structured decision-making; social network.

show abstract

“…The modern development of online social networks and phenomena like the spread of misinformation and disinformation represent widely researched topics [11–15]. Other important extensions to original epidemic models have been the inclusion of adaptive agents, which allows for richer dynamics, multi-agent models [16] based on local rules that introduce self-organizing behaviors [17], or models with game-theoretic agents that simulate strategic behaviors [18].…”

Section: Introductionmentioning

confidence: 99%

The coevolution of contagion and behavior with increasing and decreasing awareness

2019

View full text Add to dashboard Cite

Understanding the effects of individual awareness on epidemic phenomena is important to comprehend the coevolving system dynamic, to improve forecasting, and to better evaluate the outcome of possible interventions. In previous models of epidemics on social networks, individual awareness has often been approximated as a generic personal trait that depends on social reinforcement, and used to introduce variability in state transition probabilities. A novelty of this work is to assume that individual awareness is a function of several contributing factors pooled together, different by nature and dynamics, and to study it for different epidemic categories. This way, our model still has awareness as the core attribute that may change state transition probabilities. Another contribution is to study positive and negative variations of awareness, in a contagion-behavior model. Imitation is the key mechanism that we model for manipulating awareness, under different network settings and assumptions, in particular regarding the degree of intentionality that individuals may exhibit in spreading an epidemic. Three epidemic categories are considered—disease, addiction, and rumor—to discuss different imitation mechanisms and degree of intentionality. We assume a population with a heterogeneous distribution of awareness and different response mechanisms to information gathered from the network. With simulations, we show the interplay between population and awareness factors producing a distribution of state transition probabilities and analyze how different network and epidemic configurations modify transmission patterns.

show abstract

Diffusion in Social Networks: A Multiagent Perspective

Cited by 105 publications

References 101 publications

Fast consensus in a large-scale multi-agent system with directed graphs using time-delayed measurements

Fast consensus in a large-scale multi-agent system with directed graphs using time-delayed measurements

Structured Multi-Agent-Based Model for Bankruptcy Contagion With Cash Flow

The coevolution of contagion and behavior with increasing and decreasing awareness

Contact Info

Product

Resources

About