scholarly journals Penalized homophily latent space models for directed scale-free networks

PLoS ONE ◽  
2021 ◽  
Vol 16 (8) ◽  
pp. e0253873
Author(s):  
Hanxuan Yang ◽  
Wei Xiong ◽  
Xueliang Zhang ◽  
Kai Wang ◽  
Maozai Tian
Keyword(s):  
Social Networks ◽  
Online Social Networks ◽  
Distance Measure ◽  
Real Data ◽  
Directed Network ◽  
Finite Sample ◽  
Scale Free ◽  
Latent Space ◽  
Scale Free Networks ◽  
Latent Space Models

Online social networks like Twitter and Facebook are among the most popular sites on the Internet. Most online social networks involve some specific features, including reciprocity, transitivity and degree heterogeneity. Such networks are so called scale-free networks and have drawn lots of attention in research. The aim of this paper is to develop a novel methodology for directed network embedding within the latent space model (LSM) framework. It is known, the link probability between two individuals may increase as the features of each become similar, which is referred to as homophily attributes. To this end, penalized pair-specific attributes, acting as a distance measure, are introduced to provide with more powerful interpretation and improve link prediction accuracy, named penalized homophily latent space models (PHLSM). The proposed models also involve in-degree heterogeneity of directed scale-free networks by embedding with the popularity scales. We also introduce LASSO-based PHLSM to produce an accurate and sparse model for high-dimensional covariates. We make Bayesian inference using MCMC algorithms. The finite sample performance of the proposed models is evaluated by three benchmark simulation datasets and two real data examples. Our methods are competitive and interpretable, they outperform existing approaches for fitting directed networks.

scholarly journals Online Social Networks (OSN) Evolution Model Based on Homophily and Preferential Attachment

Symmetry ◽  
2018 ◽  
Vol 10 (11) ◽  
pp. 654 ◽  
Author(s):  
Jebran Khan ◽  
Sungchang Lee
Keyword(s):  
Social Networks ◽  
Structural Properties ◽  
Online Social Networks ◽  
Large Scale ◽  
Preferential Attachment ◽  
Real Life ◽  
Synthetic Data ◽  
Evolution Model ◽  
Scale Invariant ◽  
Scale Free

In this paper, we propose a new scale-free social networks (SNs) evolution model that is based on homophily combined with preferential attachments. Our model enables the SN researchers to generate SN synthetic data for the evaluation of multi-facet SN models that are dependent on users’ attributes and similarities. Homophily is one of the key factors for interactive relationship formation in SN. The synthetic graph generated by our model is scale-invariant and has symmetric relationships. The model is dynamic and sustainable to changes in input parameters, such as number of nodes and nodes’ attributes, by conserving its structural properties. Simulation and evaluation of models for large-scale SN applications need large datasets. One way to get SN data is to generate synthetic data by using SN evolution models. Various SN evolution models are proposed to approximate the real-life SN graphs in previous research. These models are based on SN structural properties such as preferential attachment. The data generated by these models is suitable to evaluate SN models that are structure dependent but not suitable to evaluate models which depend on the SN users’ attributes and similarities. In our proposed model, users’ attributes and similarities are utilized to synthesize SN graphs. We evaluated the resultant synthetic graph by analyzing its structural properties. In addition, we validated our model by comparing its measures with the publicly available real-life SN datasets and previous SN evolution models. Simulation results show our resultant graph to be a close representation of real-life SN graphs with users’ attributes.

scholarly journals COOPERATION ON SOCIAL NETWORKS AND ITS ROBUSTNESS

2012 ◽  
Vol 15 (supp01) ◽  
pp. 1250046 ◽  
Author(s):  
ALBERTO ANTONIONI ◽  
MARCO TOMASSINI
Keyword(s):  
Social Networks ◽  
Population Dynamics ◽  
Numerical Simulations ◽  
Population Model ◽  
Evolutionary Games ◽  
Computer Models ◽  
Scale Free ◽  
Average Payoff ◽  
Scale Free Networks ◽  
Update Rules

In this work we have used computer models of social-like networks to show by extensive numerical simulations that cooperation in evolutionary games can emerge and be stable on this class of networks. The amounts of cooperation reached are at least as much as in scale-free networks but here the population model is more realistic. Cooperation is robust with respect to different strategy update rules, population dynamics, and payoff computation. Only when straight average payoff is used or there is high strategy or network noise does cooperation decrease in all games and disappear in the Prisoner's Dilemma.

scholarly journals Emergence of Scale-Free Close-Knit Friendship Structure in Online Social Networks

PLoS ONE ◽  
2012 ◽  
Vol 7 (12) ◽  
pp. e50702 ◽  
Author(s):  
Ai-Xiang Cui ◽  
Zi-Ke Zhang ◽  
Ming Tang ◽  
Pak Ming Hui ◽  
Yan Fu
Keyword(s):  
Social Networks ◽  
Online Social Networks ◽  
Scale Free

Identification of Influential Online Social Network Users Based on Multi-Features

2016 ◽  
Vol 30 (06) ◽  
pp. 1659015 ◽  
Author(s):  
Qindong Sun ◽  
Nan Wang ◽  
Yadong Zhou ◽  
Zuomin Luo
Keyword(s):  
Social Networks ◽  
Online Social Networks ◽  
Online Social Network ◽  
Real Data ◽  
User Profiles ◽  
Sina Weibo ◽  
Pagerank Algorithm ◽  
Model Based ◽  
User Influence ◽  
Influential Users

The problem of discovering influential users is important to understand and analyze online social networks. The user profiles and interactions between users are significant features to evaluate the user influence. As these features are heterogeneous, it is challengeable to take all of them into a proper model for influence evaluation. In this paper, we propose a model based on personal user features and the adjacent factor to discover influential users in online social networks. Through taking the advantages of Bayesian network and chain principle of PageRank algorithm, the features of the user profiles and interactions are integratedly considered in our model. Based on real data from Sina Weibo data and multiple evaluation metrics of retweet count, tweet count, follower count, etc., the experimental results show that influential users identified by our model are more powerful than the ones identified by single indicator methods and PageRank-based methods.

scholarly journals COMPARISON OF ISING MAGNET ON DIRECTED VERSUS UNDIRECTED ERDÖS–RÉNYI AND SCALE-FREE NETWORKS

2007 ◽  
Vol 18 (01) ◽  
pp. 53-60 ◽  
Author(s):  
M. A. SUMOUR ◽  
A. H. EL-ASTAL ◽  
F. W. S. LIMA ◽  
M. M. SHABAT ◽  
H. M. KHALIL
Keyword(s):  
Spontaneous Magnetization ◽  
The Other ◽  
Directed Network ◽  
Equilibration Time ◽  
Arrhenius Law ◽  
Power Law Distribution ◽  
Scale Free ◽  
Scale Free Networks ◽  
Low Probability ◽  
Ising Magnet

Scale-free networks are a recently developed approach to model the interactions found in complex natural and man-made systems. Such networks exhibit a power-law distribution of node link (degree) frequencies n(k) in which a small number of highly connected nodes predominate over a much greater number of sparsely connected ones. In contrast, in an Erdös–Renyi network each of N sites is connected to every site with a low probability p (of the order of 1/N). Then the number k of neighbors will fluctuate according to a Poisson distribution. One can instead assume that each site selects exactly k neighbors among the other sites. Here we compare in both cases the usual network with the directed network, when site A selects site B as a neighbor, and then B influences A but A does not influence B. As we change from undirected to directed scale-free networks, the spontaneous magnetization vanishes after an equilibration time following an Arrhenius law, while the directed ER networks have a positive Curie temperature.

scholarly journals ADVERTISING AND IRREVERSIBLE OPINION SPREADING IN COMPLEX SOCIAL NETWORKS

2009 ◽  
Vol 20 (06) ◽  
pp. 799-815 ◽  
Author(s):  
JULIÁN CANDIA
Keyword(s):  
Social Networks ◽  
Small World ◽  
Cost Effective ◽  
Thermal Bath ◽  
Scale Free ◽  
Effective Strategies ◽  
Scale Free Networks ◽  
Variable Duration ◽  
Link Density ◽  
Spin Flips

Irreversible opinion spreading phenomena are studied on small-world and scale-free networks by means of the magnetic Eden model, a nonequilibrium kinetic model for the growth of binary mixtures in contact with a thermal bath. In this model, the opinion of an individual is affected by those of their acquaintances, but opinion changes (analogous to spin flips in an Ising-like model) are not allowed. We focus on the influence of advertising, which is represented by external magnetic fields. The interplay and competition between temperature and fields lead to order–disorder transitions, which are found to also depend on the link density and the topology of the complex network substrate. The effects of advertising campaigns with variable duration, as well as the best cost-effective strategies to achieve consensus within different scenarios, are also discussed.

A Parallel Neural Network Approach for Faster Rumor Identification in Online Social Networks

2019 ◽  
Vol 15 (4) ◽  
pp. 69-89 ◽  
Author(s):  
Santhoshkumar Srinivasan ◽  
Dhinesh Babu L D
Keyword(s):  
Neural Network ◽  
Social Networks ◽  
Online Social Networks ◽  
Activation Function ◽  
Network Approach ◽  
Neural Network Approach ◽  
Scale Free ◽  
Rumor Detection ◽  
Available Information ◽  
And Control

The unprecedented scale of rumor propagation in online social networks urges the necessity of faster rumor identification and control. The identification of rumors in the inception itself is imperative to bring down the harm it could cause to the society at large. But, the available information regarding rumors in inception stages is minimal. To identify rumors with data sparsity, we have proposed a twofold convolutional neural network approach with a new activation function which generalizes faster with higher accuracy. The proposed approach utilizes prominent features such as temporal and content for the classification. This rumor detection method is compared with the state-of-the-art rumor detection approaches and results prove the proposed method identifies rumor earlier than other approaches. Using this approach, the detected rumors with 88% accuracy and 92% precision for experimental datasets is 5% to 35% better than the existing approaches. This automated approach provides better results for larger and scale-free networks.

Sign in / Sign up

Export Citation Format

Share Document