Online Social Networking Platforms Research Articles

A Federated Social Recommendation Approach with Enhanced Hypergraph Neural Network

In recent years, the development of online social network platforms has led to increased research efforts in social recommendation systems. Unlike traditional recommendation systems, social recommendation systems utilize both user-item interactions and user-user social relations to recommend relevant items, taking into account social homophily and social influence. Graph neural network (GNN) based social recommendation methods have been proposed to model these item interactions and social relations effectively. However, existing GNN-based methods rely on centralized training, which raises privacy concerns and faces challenges in data collection due to regulations and privacy restrictions. Federated learning has emerged as a privacy-preserving alternative. Combining federated learning with GNN-based methods for social recommendation can leverage their respective advantages, but it also introduces new challenges: 1) existing federated recommendation systems often lack the capability to process heterogeneous data, such as user-item interactions and social relations; 2) due to the sparsity of data distributed across different clients, capturing the higher-order relationship information among users becomes challenging and is often overlooked by most federated recommendation systems. To overcome these challenges, we propose a federated social recommendation approach with enhanced hypergraph neural network. We introduce hypergraph graph neural networks (HGNN) to learn user and item embeddings in federated recommendation systems, leveraging the hypergraph structure to address the heterogeneity of data. Based on carefully crafted triangular motifs, we merge user and item nodes to construct hypergraphs on local clients, capturing specific triangular relations. Multiple HGNN channels are used to encode different categories of high-order relations, and an attention mechanism is applied to aggregate the embedded information from these channels. Our experiments on real-world social recommendation datasets demonstrate the effectiveness of the proposed approach. Extensive experiment results on three publicly available datasets validate the effectiveness of the proposed method.

Modeling the dissemination of privacy information in online social networks

AbstractDriven by the rapid development of information technology, online social networks (OSNs) have experienced a fast development in recent years, allowing increasingly more people to share and spread information over OSNs. The rapid rise of OSN platforms such as Facebook and Twitter is sufficient evidence of such development. As one type of information, privacy information can also be created and disseminated over an OSN, posing a severe threat to individual privacy. This article attempts to construct a model for disseminating privacy information in OSNs and to analyze the model by simulating the dissemination process of privacy information in OSNs. First, we establish network models that exhibit the main characteristics of OSNs. Second, by considering the factors related to social relationships, especially intimacy between users and the attention of users to the privacy subject, we derive the parameters for privacy information dissemination models in OSNs. Third, based on the theory of information dissemination dynamics, we construct a model for information dissemination that conforms to the properties of privacy information. We also present some experimental results based on the constructed model and analyze the characteristics of privacy information dissemination. Fourth, we study and verify the various properties of the model through a set of experiments. The proposed model provides the opportunity to better understand the dynamics of privacy information dissemination in OSNs and the effect of user behavior on dissemination.

Detection of Fake Profiles on Online Social Network Platforms: Performance Evaluation of Artificial Intelligence Techniques

Detection of Fake Profiles on Online Social Network Platforms: Performance Evaluation of Artificial Intelligence Techniques

A novel Cosine-Cosine chaotic map-based video encryption scheme

The surge in online activities has led to the increasing popularity of sharing video data across diverse applications, including online education tutorials, social networking, video calling, and OTT platforms. Encryption prevents unauthorized access to the transmitted data over unreliable channels. The well-known features of chaos theory such as random behaviour, unpredictability, and initial parameters dependency facilitate its use in cryptography. Many security issues are faced by chaos-based cryptosystems because of their less complexity. Hence, a new Cosine-Cosine chaotic map characterized by intricate chaotic behaviour is designed in the current study. Additionally, we formulate an original video encryption scheme employing this Cosine-Cosine chaotic map. The encryption process involves five steps, beginning with the segmentation of the original video into frames based on its frame rate. In the second phase, a 384 bits pseudorandom key is generated that is further divided into three subkeys of 128 bits each. The novel Cosine-Cosine chaotic map-based sequence is generated. In the fourth step, red, green, and blue components are encrypted using the pseudorandom key and the chaotic sequence. In the last step, we combine encrypted frames to get cipher video. The security analysis validates that the proposed encryption protects against eavesdropping.

Game-theoretic modeling and analysis of cyberbullying spreading on OSNs

The rapid development of online social networks (OSNs) has enabled seamless cyberbullying activities, which not only cause psychological harm to victims but also bring disharmony to society. From the perspective of OSN platform administrators, strategic decision making towards when/how to launch an anti-cyberbullying campaign is the key to mitigating its negative impact. Hence, this paper develops a dynamic anti-cyberbullying investment solution to mitigate the adverse impact of strategic cyberbullying phenomena cost-effectively. Specifically, this paper first quantifies the payoffs of the attacker/bullying instigator and the defender/platform administrator based on a newly proposed propagation model. Then, considering the budget-constrained factor for both sides, a differential game-theoretical model is proposed to simulate each party's optimal trade-off between cost and benefit. Next, by utilizing Pontryagin Maximum/Minimum Principle, a set of necessary conditions for a Nash equilibrium strategy profile have been proposed. Based on the forward-backward sweep method, an iterative algorithm (SCR optimization algorithm) is tailored to solve the conditions numerically. Through experiments conducted under Facebook, Youtube and Twitter sub-networks, the algorithm's performance has been evaluated by inspecting its convergence speed and payoff compared with randomized profiles. Consequently, the resulting strategies are verified to exhibit superior performance in dealing with cyberbullying spreading.

Toxic Fake News Detection and Classification for Combating COVID-19 Misinformation

The emergence of COVID-19 has led to a surge in fake news on social media, with toxic fake news having adverse effects on individuals, society, and governments. Detecting toxic fake news is crucial, but little prior research has been done in this area. This study aims to address this gap and identify toxic fake news to save time spent on examining nontoxic fake news. To achieve this, multiple datasets were collected from different online social networking platforms such as Facebook and Twitter. The latest samples were obtained by collecting data based on the topmost keywords extracted from the existing datasets. The instances were then labeled as toxic/nontoxic using toxicity analysis, and traditional machine-learning (ML) techniques such as linear support vector machine (SVM), conventional random forest (RF), and transformer-based ML techniques such as bidirectional encoder representations from transformers (BERT) were employed to design a toxic-fake news detection (FND) and classification system. As per the experiments, the linear SVM method outperformed BERT SVM, RF, and BERT RF with an accuracy of 92% and -score, -score, and -score of 95%, 85%, and 87%, respectively. Upon comparison, the proposed approach has either suppressed or achieved results very close to the state-of-the-art techniques in the literature by recording the best values on performance metrics such as accuracy, F1-score, precision, and recall for linear SVM. Overall, the proposed methods have shown promising results and urge further research to restrain toxic fake news. In contrast to prior research, the presented methodology leverages toxicity-oriented attributes and BERT-based sequence representations to discern toxic counterfeit news articles from nontoxic ones across social media platforms.

How tobacco companies use the revolving door between government and industry to influence policymaking: an Australian case study.

Mixed-methods including non-experimental descriptive and exploratory case studies. To build a picture of tobacco lobbying through the revolving door in Australia, data was triangulated from multiple publicly available sources: 1) Australian federal, state and territory government lobbyist registers, 2) the online social networking platform, LinkedIn; and 3) Australian news media reports. Tobacco companies lobby the Australian government using 'in-house' employees, lobbyists working in firms acting on their behalf, and third-party allies with common interests. Almost half (48%) of internal tobacco company lobbyists had held positions in the Australian government (state, territory and/or federal) before or after working in the tobacco industry. Likewise, 55% of lobbyists acting on behalf of tobacco companies had held government positions before or after working as a lobbyist. In-house tobacco industry lobbyists, as well as those working on behalf of tobacco companies within lobbying firms, were found to have held senior governmental positions, such as a Member of Parliament (MP) or Senator, chief or deputy chief of staff, or senior advisor in a ministerial office, and many had moved into or out of government within 1 year of working for a tobacco company (56%) or as a lobbyist (48%). Tobacco companies are strategically using the revolving door between the government and the tobacco industry as a key political lobbying mechanism to try to influence public health policy in Australia.

CALEB: A Conditional Adversarial Learning Framework to enhance bot detection

The high growth of Online Social Networks (OSNs) over the last few years has allowed automated accounts, known as social bots, to gain ground. As highlighted by other researchers, many of these bots have malicious purposes and tend to mimic human behavior, posing high-level security threats on OSN platforms. Moreover, recent studies have shown that social bots evolve over time by reforming and reinventing unforeseen and sophisticated characteristics, making them capable of evading the current machine learning state-of-the-art bot detection systems. This work is motivated by the critical need to establish adaptive bot detection methods in order to proactively capture unseen evolved bots towards healthier OSNs interactions. In contrast with most earlier supervised ML approaches which are limited by the inability to effectively detect new types of bots, this paper proposes CALEB, a robust end-to-end proactive framework based on the Conditional Generative Adversarial Network (CGAN) and its extension, Auxiliary Classifier GAN (AC-GAN), to simulate bot evolution by creating realistic synthetic instances of different bot types. These simulated evolved bots augment existing bot datasets and therefore enhance the detection of emerging generations of bots before they even appear. Furthermore, we show that our augmentation approach overpasses other earlier augmentation techniques which fail at simulating evolving bots. Extensive experimentation on well established public bot datasets, show that our approach offers a performance boost of up to 10% regarding the detection of new unseen bots. Finally, the use of the AC-GAN Discriminator as a bot detector, has outperformed former ML approaches, showcasing the efficiency of our end to end framework.

Techniques for predicting dark web events focused on the delivery of illicit products and ordered crime

Malicious actors, specially trained professionals operating anonymously on the dark web (DW) platform to conduct cyber fraud, illegal drug supply, online kidnapping orders, CryptoLocker induction, contract hacking, terrorist recruitment portals on the online social network (OSN) platform, and financing are always a possibility in the hyperspace. The amount and variety of unlawful actions are increasing, which has prompted law enforcement (LE) agencies to develop efficient prevention tactics. In the current atmosphere of rapidly expanding cybercrime, conventional crime-solving methods are unable to produce results due to their slowness and inefficiency. The methods for accurately predicting crime before it happens "automated machine" to help police officers ease the burden on personnel while also assisting in preventing offense. To achieve and explain the results of a few cases in which such approaches were applied, we advise combining machine learning (ML) with computer vision (CV) strategies. This study's objective is to present dark web crime statistics and a forecasting model for generating alerts of illegal operations like drug supply, people smuggling, terrorist staffing and radicalization, and deceitful activities that are connected to gangs or organizations showing online presence using ML and CV to help law enforcement organizations identify, and accumulate proactive tactics for solving crimes.

DeepMUI: A novel method to identify malicious users on online social network platforms

SummaryThe use of online social network (OSN) platforms has become an essential component of contemporary society, facilitating global connectivity, and information sharing among individuals. The proliferation of malicious users has emerged as a noteworthy obstacle, exerting a detrimental effect on the authenticity of the data disseminated through these channels. A malicious profile is created with the intention of disseminating false information, manipulating perspectives, and executing harmful actions, including phishing schemes, identity theft, and the propagation of malware. Consequently, the identification of malicious users has emerged as an essential undertaking for both OSN platforms and researchers. The objective of this study is to investigate the issue of identifying malicious users on OSN platforms. The DeepMUI model has been introduced as a new approach to identifying malicious users on OSN platforms, utilizing user profile metadata‐derived characteristics. The DeepMUI architecture is composed of long short‐term memory and convolutional neural network models. Additionally, it integrates alterations to the pooling layer to improve its overall efficacy. The experiments have demonstrated that DeepMUI exhibits promising results in the task of identifying malicious users, with greater accuracy and minimal loss compared to existing methods.

Depolarization of opinions on social networks through random nudges.

Polarization of opinions has been empirically noted in many online social network platforms. Traditional models of opinion dynamics, based on statistical physics principles, do not account for the emergence of polarization and echo chambers in online network platforms. A recently introduced opinion dynamics model that incorporates the homophily factor-the tendency of agents to connect with those holding similar opinions as their own-captures polarization and echo chamber effects. In this work, we provide a nonintrusive framework for mildly nudging agents in an online community to form random connections. This is shown to lead to significant depolarization of opinions and decrease the echo chamber effects. Though a mild nudge effectively avoids polarization, overdoing this leads to another undesirable effect, namely, radicalization. Further, we obtain the optimal nudge probability to avoid the extremes of polarization and radicalization outcomes.

Go-Sharing: A Blockchain-Based Privacy-Preserving Framework for Cross-Social Network Photo Sharing

The evolution of social media has led to a trend of posting daily photos on online Social Network Platforms (SNPs). The privacy of online photos is often protected carefully by security mechanisms. However, these mechanisms will lose effectiveness when someone spreads the photos to other platforms. In this paper, we propose Go-sharing, a blockchain-based privacy-preserving framework that provides powerful dissemination control for cross-SNP photo sharing. In contrast to security mechanisms running separately in centralized servers that do not trust each other, our framework achieves consistent consensus on photo dissemination control through carefully designed smart contract-based protocols. We use these protocols to create platform-free dissemination trees for every image, providing users with complete sharing control and privacy protection. Considering the possible privacy conflicts between owners and subsequent re-posters in cross-SNP sharing, we design a dynamic privacy policy generation algorithm that maximizes the flexibility of re-posters without violating formers' privacy. Moreover, Go-sharing also provides robust photo ownership identification mechanisms to avoid illegal reprinting. It introduces a random noise black box in a two-stage separable deep learning process to improve robustness against unpredictable manipulations. Through extensive real-world simulations, the results demonstrate the capability and effectiveness of the framework across a number of performance metrics.

Text Mining of Twitter data for Mapping the Digital Humanities Research Trends

Digital humanities have become a more relevant field of study due to the extraordinary growth in digitisation of the humanities data. Due to collaborative development of humanities and computing, many academics are convinced of the worth of digital humanities (DH) that actually provides the best insight into humanities studies. The panoramic view of the development of big data in humanities reflects its trendy directions and evoked new challenges in DH. It is complicated to analysed the objectives of digital humanities data with simple data analysis tools where as text mining can help to facilitate the qualitative findings in DH. In the humanities disciplines, data is often in the form of unstructured and text mining is a way of structuring and analysing digitised text-as-data. Twitter is a online social networking platform which offers an opportunity for quality information sharing, collaborative participation digital humanities community. This paper is attempted to study the extensibility of digital humanities on twitter and also to interpret the evolution of twitter usage by analysing tweets posted related to DH via python data analysis.

Depression Screening in Humans With AI and Deep Learning Techniques

Social media platforms have been widely used as a communication tool where most of the population expresses their feelings and shares life experiences. Along with general information about the public, these platforms hold an ample amount of content related to depressed users and thus can generate sensitive social signals indicating if a person is suffering from some serious issues, such as self-harm, suicidal thoughts, or intention for an unlawful act. Early depression detection using advanced natural language processing (NLP), deep machine learning, and transfer learning techniques can assist in designing an efficient system to detect major depressive systems at an early stage. The current depression detection models are not enough to capture sensitive social signals indicating the true mood, personality, and behavior of an individual. Thus, making the current systems unsatisfactory. To address this life-threatening human-health problem, we propose an efficient artificial intelligence (AI) and deep learning (DL)-based model for identifying depressed individuals on social media platforms. The model employs hybrid feature-based behavioral-biometric signals captured using Word2Vec, term frequency-inverse document frequency (TF-IDF) models to learn a convolutional neural network (CNN) and long-short term memory (LSTM) models. The data are captured from multiple sources using advanced crawling strategies to have data variety in the corpus. Thus, making the proposed system effective across platforms. The Dataset produced by this study is the first of its kind with a variety of depressive signals from online social network (OSN) platforms including Facebook, Twitter, and YouTube. The experiments have shown that both DL models LSTM and CNN, and the hybrid (CNN <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$+$</tex-math> </inline-formula> LSTM) models achieved promising results on all individual as well as combined datasets. Out of 24 experiments for Word2Vec LSTM and Word2Vec (CNN <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$+$</tex-math> </inline-formula> LSTM) models, we achieved the accuracy of 99.02% and 99.01%, respectively, and recorded as best results outperforming all the existing approaches on performance measures such as recall, precision, accuracy, and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">F</i> 1-score. The Word2Vec-based features have been proved optimal features for detecting depressions symptoms on Facebook corpus (FC) and YouTube corpus (YC) by achieving an accuracy of 95.02% (with CNN) and 98.15% (with CNN <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$+$</tex-math> </inline-formula> LSTM), respectively.

BBC-FND: An ensemble of deep learning framework for textual fake news detection

A wide spread of false news over Online Social Network platforms (OSNs) causes numerous negative consequences. Several researchers proposed different models using machine learning (ML) and deep learning (DL) techniques to detect fake news. The performance analysis of these models exposed that they achieve low accuracy due to the major drawbacks like improper architectural design, unsuitability of the model on different datasets. To address these issues, a hybrid BERT-BiLSTM-CNN (BBC-FND) model is proposed. It comprises of three main layers: an embedding layer, a feature representation layer, and a classification layer. In the first layer, BERT is used to extract the contextual-dependent features among words from the news. A multichannel CNN and a stacked BiLSTM are employed to produce various key features in the feature extraction layer. The former extracts multiple features from the text and captures complex local patterns in word spatial relations, while the latter extracts global semantic features from the contextual feature vector. The resulting features are concatenated and fed as input to the FFN to predict the news as fake or real. The results exhibit that the BBC-FND model outperforms the SoTA techniques, achieving higher accuracy of 97.31%, 98.64%, 99.06% and 98.26% on four datasets respectively.

Public perception of electric vehicles on Reddit and Twitter: A cross-platform analysis

Electrified mobility such as electric vehicles (EVs) is a promising solution to reduce carbon emissions in transportation and mitigate global warming. Understanding public perception of EVs can help better support their adoption. A previous study shows that online social networks (OSNs) such as Reddit can be a valuable source for studying public perceptions of EVs and provide different perspectives from traditional methods that leverage surveys, questionnaires, or interviews (Ruan and Lv, 2022). Our work aims to investigate this direction further through the following research question: Given the distinct mechanisms of various OSNs, can we obtain a more comprehensive picture of public perception of EVs by integrating the analysis from different platforms? Specifically, our study is based on EV-related discussions on two popular OSN platforms: Twitter and Reddit. We have collected 3,437,917 Reddit posts (including 274,979 submissions and 3,162,938 comments) and 7,383,327 Tweets between January 2011 and December 2020 and analyzed them from several perspectives. Our analysis shows that users have had different topic and sentiment patterns in EV-related discussions on the two platforms over the past decade. We also leverage the verified account information on Twitter to reveal that the most influential users are politicians and news media; however, the general public has very different conversation patterns with the two types of accounts — politicians seem to be increasingly (over) optimistic about EVs while the public may think differently.

Image Sanitization in Online Social Networks: A General Framework for Breaking Robust Information Hiding

With the development of robust information hiding (RIH) approaches, secret messages can be extracted successfully from stego-data after transmission through lossy channels of online social networks (OSNs). To interrupt illegal covert communications in OSNs, some methods sanitize the uploaded images by image processing operations to destroy the hidden data that may exist. However, none of the existing methods takes the RIH methods that can resist scaling into consideration, while scaling is a common operation in OSNs. In this paper, we first propose a general framework for image sanitization in OSN platforms, which serves as a countermeasure against the RIH. By using such a framework, the secret messages embedded in the upload images can be removed and the quality of the sanitized image can be well maintained. Our framework contains two deep neural networks: Scaling-Net and SC-Net. The Scaling-Net is dedicated to the sanitization of oversized images while the SC-Net is designed for other images. To achieve a good image quality, we also propose a discriminator for adversarial training of the Scaling-Net and SC-Net. Experimental results on different datasets demonstrate that our proposed method outperforms the state-of-the-art methods. The source code and pretrained models are available at our code repository <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> .

AN OBSERVATIONAL STUDY OF PRACTICE OF COVID-19 VACCINATION AMONG HEALTHCARE WORKERS AT A TERTIARY HEALTHCARE CENTRE: A SURVEY BASED ANALYSIS.

Background: Coronavirus disease 2019 (COVID-19) is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The rapid and extensive spread of the COVID-19 pandemic has become a major cause of concern for the healthcare profession. Development of SARS-CoV-2 vaccine is considered as the only way to signicantly reduce the threat of this pandemic in view of no proper treatment available against COVID-19 till date. However, the success of COVID-19 vaccination in India is dependent upon the acceptance of vaccine among general population, besides the proven safety and efcacy of the Indian COVID-19 vaccines (Co vishield and Co vaxin). In view of that, we conducted an online survey among the rst to be vaccinated priority group (healthcare workers) in India who are at high risk of exposure, with the aim to assess the beliefs and experiences related to Indian COVID-19 vaccines and to project that upon the general population of India. Method: This cross sectional study was conducted using a validated self administered online questionnaire that was distributed through emails, online social networking platforms (online survey) and also printed form of questionnaire in both English and Hindi languages (paper survey). The participants of the study were informed about the objectives of the study and after obtaining an informed consent, they completed the questionnaire on their background and vaccinationbehaviour related variables (including past vaccine compliance, risk factors for severe COVID-19, and COVID-19 vaccine perceptions and experience). Results: Our study included 750 healthcare workers (HCW) of which 415 were doctors, 286 were medical students and 49 were paramedical staff, Of the total number of participants, n=617 (82.3%) have received the COVID-19 vaccine. 75.6% of the study population (n=504) were not hesitant for getting vaccination against COVID-19. Majority of the HCWs believed in the safety and efcacy of COVID-19 vaccine and wanted their family members to get vaccinated. Conclusions: Our study concludes that Health Care Workers show acceptance and condence in the vaccination programme against Covid 19. This can contribute to general population compliance and willingness to accept Covid 19 vaccine.

HE-SNA: an efficient cross-platform network alignment scheme from privacy-aware perspective

User alignment across online social network platforms (OSNPs) is a growing concern with the rapid development of internet technology. In reality, users tend to register different accounts on multiple OSNPs, and the network platforms are reluctant to share network structure and user’s information due to business interest and privacy protection, which brings great obstacles to cross-platform user alignment. In view of this, we propose a homomorphic encryption-based social network alignment (HE-SNA) algorithm from the perspective of privacy leakage. Specifically, we first consider the OSNPs as a system containing multiple social networks, that each participant of OSNPs owns part of the network, i.e., a separate private sub-network. Then, encryption, fusion and decryption operations of the alignment information are performed by two third-party servers using HE scheme, which can protect the privacy information of sub-networks effectively. Finally, each sub-network uses the fused alignment information sent back from the third-party server for user alignment. Experimental results show that the HE-SNA method can provide a sum of locally trained models to third-party servers without leaking the privacy of any single sub-network. Moreover, the HE-SNA achieves a promising network alignment performance than only using the structural information and alignment data of single private sub-network while protecting its topology structure information.

Mitigating Influence of Disinformation Propagation Using Uncertainty-Based Opinion Interactions

For decades, the spread of disinformation in online social networks (OSNs) has been a serious social issue. Disinformation via social media can easily mislead people’s beliefs toward or against an event that may mislead their behaviors based on the misbeliefs. The game theory approaches have been proposed under dynamic settings to limit the adverse influences of disinformation. It is a challenge to expand the users’ game strategies from the spreading decisions to the possible opinion updating choices. This work proposes a game-theoretic opinion framework that can formulate dynamic opinions by a belief model called Subjective Logic (SL) and provide opinion updates on five types of users’ interactions on OSN platforms. The opinions are updated based on user choices and user types through the game interactions among legitimate users, attackers, and a defender in an OSN. Via the extensive simulation experiments, the effectiveness of the opinion models of five decision-makers (DMs) is analyzed in terms of users believing or disbelieving disinformation in an epidemic model with parameter optimization. Our results show that while homophily-based DMs (H-DMs) introduce the highest opinion polarization, uncertainty-based DMs (U-DMs) can effectively filter untrustworthy users propagating disinformation.