Article
Computer Science, Information Systems
Sijing Chen, Lu Xiao, Jin Mao
Summary: Social media users are easily persuaded by misinformation and contribute to its propagation. Pathos strategies are found to be the most common persuasion tactic in misinformation-containing posts, and the type of persuasion strategies used correlate with the topics, expected actions, and digital elements of the posts, affecting users' responses significantly. This study sheds light on the complex and context-dependent mechanisms behind the presentation and propagation of misinformation in social media.
INFORMATION PROCESSING & MANAGEMENT
(2021)
Article
Mathematics, Interdisciplinary Applications
Lilei Han, Zhaohua Lin, Qingqing Yin, Ming Tang, Shuguang Guan, Marian Boguna
Summary: This paper proposes a general formalism to study non-Markovian dynamics on non-Markovian temporal networks. The study finds that, under certain conditions, non-Markovian dynamics on temporal networks are equivalent to Markovian dynamics on static networks, independent of the underlying network topology.
CHAOS SOLITONS & FRACTALS
(2023)
Article
Computer Science, Theory & Methods
Daniele Bellutta, Kathleen M. Carley
Summary: Democracies worldwide are threatened by coordinated online influence campaigns that manipulate their electorates. Researchers have developed valuable methods to detect automated accounts and identify false information, but struggle to keep up with the constantly changing behavior of the perpetrators. By analyzing tweets discussing the 2020 U.S. elections, it was found that sudden increases in Twitter account creations could serve as early warnings of online information operations. These accounts displayed similar behavior, agreement on certain topics, higher likelihood of being bots, and shared links to low-credibility sites. Combined with other techniques, social media platforms could temporarily limit the influence of accounts created during these burst periods.
JOURNAL OF BIG DATA
(2023)
Article
Mathematics, Interdisciplinary Applications
Matteo Bruno, Renaud Lambiotte, Fabio Saracco
Summary: Online social networks play crucial roles in political campaigns, but they also carry potential risks such as misinformation campaigns and malicious activities. This study examines the interactions between users and bots during the UK elections in 2019, focusing on the polarized discussion about Brexit on Twitter. The findings reveal the influence of automated accounts during the days leading up to the national elections, which decreases afterwards. Additionally, the number of suspended users significantly increases after the election day. The study also explores the political orientation of the bots and their usage of hashtags and URLs to shape common narratives.
Article
Psychology, Multidisciplinary
Sijing Chen, Lu Xiao, Akit Kumar
Summary: This study aims to provide a systematic overview of the factors influencing the spread of misinformation on social media, as well as to summarize current strategies against it. Through a systematic literature review, 423 relevant articles were identified and analyzed. Research gaps were identified and future research directions were recommended.
COMPUTERS IN HUMAN BEHAVIOR
(2023)
Article
Mathematics, Applied
Yunxiang Hou, Yikang Lu, Yuting Dong, Libin Jin, Lei Shi
Summary: This study examines the impact of social attitudes on the transmission of infectious diseases in activity-driven networks. The research shows that reducing social intensity and increasing the number of risk-averse individuals are effective strategies in controlling epidemic outbreaks.
APPLIED MATHEMATICS AND COMPUTATION
(2023)
Article
Computer Science, Artificial Intelligence
Javier Huertas-Tato, Alejandro Martin, David Camacho
Summary: The emergence of complex attention-based language models like BERT, RoBERTa or GPT-3 has enabled the tackling of highly complex tasks in various scenarios. However, these models face significant difficulties when applied to specific domains, such as social networks like Twitter. In order to address the challenges of natural language processing in this domain, we present BERTuit, the largest transformer proposed for the Spanish language, pre-trained on a massive dataset of Spanish tweets. Our motivation is to provide a powerful resource for better understanding Spanish Twitter and combating the spread of misinformation. BERTuit is evaluated and compared against competitive multilingual transformers, showing its utility through applications like visualizing groups of hoaxes and profiling authors spreading disinformation.
Article
Engineering, Multidisciplinary
Marcin Waniek, Petter Holme, Talal Rahwan
Summary: Social network analysis tools can infer various attributes by examining connections. Previous studies on hiding personal importance in static networks have overlooked the more general case of temporal networks. This research investigates the concealment of personal importance in temporal networks with changing edges. The study shows that finding the optimal hiding strategy is usually computationally infeasible, but manipulating contacts can increase privacy. Temporal networks provide more strategies for manipulation compared to static networks, making hiding easier.
IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING
(2022)
Article
Mathematics, Interdisciplinary Applications
Linfeng Zhong, Yu Bai, Changjiang Liu, Juan Du, Weijun Pan
Summary: This paper investigates the dynamics of information spreading on temporal networks and finds that network memory can affect the suppression and promotion of information spreading, depending on the degree heterogeneity and fraction of bigots.
Article
Computer Science, Information Systems
Jianming Zhu, Peikun Ni, Guoqing Wang, Yuan Li
Summary: The booming development of online social media has changed the way people post and access information, leading to challenges in information authenticity. This paper studies the strategy of disbanding private groups in online social networks to reduce the spread of misinformation and presents solutions.
INFORMATION SCIENCES
(2021)
Article
Management
Mohamed Mostagir, James Siderius
Summary: This article studies the spread of misinformation in a social network with unequal access to learning resources. The study shows that inequality plays a significant role in the spread of misinformation, and the relationship between the prevalence of misinformation and inequality is nonmonotonic.
MANAGEMENT SCIENCE
(2022)
Article
Multidisciplinary Sciences
Benjamin A. Lyons, Jacob M. Montgomery, Andrew M. Guess, Brendan Nyhan, Jason Reifler
Summary: Overconfidence can lead to biased judgment of true and false news, affecting behavior and beliefs; individuals who are overconfident are more likely to visit untrustworthy websites, fail to distinguish between true and false claims, and are more willing to share false content.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2021)
Article
Computer Science, Hardware & Architecture
Michael Simpson, Venkatesh Srinivasan, Alex Thomo
Summary: This article presents a solution to the problem of misinformation propagation in social networks using the Reverse Prevention Sampling (RPS) algorithm. The algorithm is proven to be effective and efficient in both theoretical analysis and experimental evaluations.
Review
Computer Science, Hardware & Architecture
Ahmad Zareie, Rizos Sakellariou
Summary: Online social networks provide a platform for rapid dissemination of information, but misinformation can also spread quickly, leading to concerns about reliability and trust. Detecting and containing the spread of misinformation has become a key focus in social network analysis.
JOURNAL OF NETWORK AND COMPUTER APPLICATIONS
(2021)
Article
Nursing
Pamela J. Grace
Summary: Nurses are expected to provide accurate information, but some nurses have been spreading misinformation on social media. This article explores why people believe they are fully informed, and the increased ethical responsibilities of nurses to address cognitive biases and biases. Strategies for nurse leaders, managers, and educators are provided to promote good practice and ensure accountability for nurses' actions and social media posts.
AMERICAN JOURNAL OF NURSING
(2021)
Review
Engineering, Electrical & Electronic
Ramji Tangudu, Prasant Kumar Sahu
Summary: This paper reviews the current status, principle of operations, challenges, and potential applications of fiber optic DTS systems. Besides discussing the development and research in this field, commercially available DTS systems and their potential applications are also reviewed. The fiber optic DTS system shows wide potential applications in various fields.
IETE TECHNICAL REVIEW
(2022)
Article
Computer Science, Hardware & Architecture
Rui Tang, Xingshu Chen, Chuancheng Wei, Qindong Li, Wenxian Wang, Haizhou Wang, Wei Wang
Summary: This paper proposes an interlayer link prediction framework based on multiple structural attributes (MulAtt) that calculates the matching degree of unmatched nodes once by leveraging the information of closed triad, intralayer links, matched neighbors, and intralayer links of neighbors simultaneously to ensure accuracy while reducing time consumption. The framework achieves better performance than several existing network structure-based methods in a non-iterative way.
Article
Automation & Control Systems
Yulong Wang, Qixu Wang, Xingshu Chen, Dajiang Chen, Xiaojie Fang, Mingyong Yin, Ning Zhang
Summary: This article proposes a noise-resilient and real-time detection system called ContainerGuard to detect Meltdown and Spectre attacks in container-based big data platforms. ContainerGuard collects performance data of processes in containers and uses generative neural networks to learn representations of normal patterns, achieving excellent detection performance.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Article
Mathematics, Applied
Yanyi Nie, Xiaoni Zhong, Tao Lin, Wei Wang
Summary: This study proposes a mathematical model to investigate the effects of homophily on heterogeneous populations with higher-order interactions. It is found that increasing 1-simplex transmission rate contributes to the spread of both behaviors, while decreasing the proportion of ω(AB) leads to a significant decrease in the final adopted density of the system.
APPLIED MATHEMATICS AND COMPUTATION
(2022)
Article
Mathematics, Applied
Dandan Zhao, Runchao Li, Hao Peng, Ming Zhong, Wei Wang
Summary: In this study, a framework for investigating the percolation of simplicial complexes with arbitrary dimensions is developed, taking into account the effects of higher-order and pairwise interactions. The robustness of simplicial complexes is assessed and properties of the model are calculated, revealing the double transition characteristics of the system.
APPLIED MATHEMATICS AND COMPUTATION
(2022)
Article
Mathematics, Applied
Wenjie Li, Yanyi Nie, Wenyao Li, Xiaolong Chen, Sheng Su, Wei Wang
Summary: This paper proposes a competing spread model for two epidemics on higher-order networks and analyzesthe factors that affect the spread process. The experimental results show that the difference in 1-simplex infection rates between the two epidemics and the increase in 2-simplex infection rates have significant impacts on the spread process.
Article
Mathematics, Applied
Jun Wang, Shimin Cai, Wei Wang, Tao Zhou
Summary: In this paper, a novel mathematical model is proposed to study the link cooperation effect of two epidemics cooperatively spreading on complex networks. The research findings show that the link cooperation effect promotes the epidemic outbreak size, and the phase transition phenomenon is closely related to the strength of the link cooperation effect and network topology.
APPLIED MATHEMATICS AND COMPUTATION
(2023)
Article
Mathematics, Interdisciplinary Applications
Yanyi Nie, Xiaoni Zhong, Tao Lin, Wei Wang
Summary: The mutation of pathogens is the most important reason for pathogen diversity. The role of traffic networks and gene networks in shaping pathogen diversity lacks theoretical study. This study models the reaction-diffusion process of pathogens on meta-population networks and finds that traffic networks make pathogen diversity more likely in cities with lower infection densities. Star-type gene networks are more likely to lead to pathogen diversity than lattice-type and chain-type gene networks.
CHAOS SOLITONS & FRACTALS
(2023)
Article
Mathematics, Applied
Qi Li, Le Cheng, Wei Wang, Xianghua Li, Shudong Li, Peican Zhu
Summary: Influence maximization is a significant topic in social network research, with potential commercial and social value. This study proposes a novel approach called the layered gravity bridge algorithm (LGB) to address the influence maximization problem. The LGB algorithm emphasizes local structural information and combines community detection algorithms with an improved gravity model. Experimental results on practical datasets demonstrate that the proposed algorithm outperforms existing methods in terms of the number of ultimately infected nodes.
APPLIED MATHEMATICS AND COMPUTATION
(2023)
Article
Computer Science, Artificial Intelligence
Xinyu Chen, Haizhou Wang, Liang Ke, Zhipeng Lu, Hanjian Su, Xingshu Chen
Summary: To reduce the negative impacts of rumors on the real world, detecting rumors on social networks is of practical significance. While there is comprehensive research on Chinese rumor detection, Cantonese rumors have been less investigated. This study proposes a novel framework for Cantonese rumor detection using deep neural networks with feature fusion. To achieve this, a Cantonese rumor dataset and a multi-domain Cantonese corpus are built. The proposed model, BLA, integrates statistical and semantic features and achieves remarkable performance with an F1 Score of 0.9225.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Min Yang, Xingshu Chen, Liuyan Tan, Xiao Lan, Yonggang Luo
Summary: With the proliferation and overwhelming data ocean of big data technology, organizations face frequent data breaches due to inefficient data security management. Data classification has become a hot topic, particularly in China, as a means of protecting data by categorizing information types and determining appropriate protective measures. In this paper, the authors introduce GENONTO, a framework that uses machine learning and natural language processing techniques to automatically extract data classification practices from 38 real-world regulations in China. GENONTO organizes this information into a structured ontology, providing valuable guidance for data practitioners and bridging the gap between expert knowledge and practical implementation.
INFORMATION PROCESSING & MANAGEMENT
(2023)
Article
Computer Science, Information Systems
Meng Cai, Han Luo, Xiao Meng, Ying Cui, Wei Wang
Summary: When public health emergencies occur, social bots disseminate a large amount of low-credibility information, manipulating public sentiment and posing a potential threat to social media's public opinion ecology. This study explores how social bots influence the mechanism of information diffusion in social networks using machine learning and causal regression methods. The findings reveal that social bots play an important role in certain topics, predominantly transmitting information with negative sentiments, but are weaker than human users in spreading negative sentiments. The study also demonstrates the predictive relationship between sentiments of humans and bots. These results provide practical suggestions for emergency management and contribute to the identification and analysis of social bots, ensuring network security and social order stability.
INFORMATION PROCESSING & MANAGEMENT
(2023)
Article
Automation & Control Systems
Rui Tang, Zhenxiong Miao, Shuyu Jiang, Xingshu Chen, Haizhou Wang, Wei Wang
Summary: Researchers propose a framework based on multiple types of consistency to predict links between different layers in a multiplex social network. The framework leverages the consistency between embedding vectors and the positional relationships of nodes in latent spaces, modeling layers as weighted graphs. Experimental results demonstrate that the framework achieves high accuracy in link prediction.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Computer Science, Cybernetics
Zhenxiong Miao, Xingshu Chen, Haizhou Wang, Rui Tang, Zhou Yang, Tiemai Huang, Wenyi Tang
Summary: The prevalence of offensive language on social networks has had negative effects on society, including online abuse. Detecting offensive language and preventing its spread is urgent. Current datasets have imbalanced distributions of users and tweets, limiting model generalization. Research has shown that incorporating community information from social graphs can improve offensive language detection, but existing models treat social graphs independently, which hinders their effectiveness. In this article, we introduce a new dataset with users and social relationships. We construct social graphs based on user behavior and relationships to encode community information. Additionally, we propose a model called GF-OLD, which uses graph attention networks (GATs) and fusion features for offensive language detection. Our approach outperforms baselines with an F1-score of 89.94%. These results demonstrate that our model effectively learns valuable information from social graphs and text, with user behavior information being particularly useful for social graph attributes.
IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS
(2023)
