Relationship Privacy Preservation in Publishing Online Social Networks

The online social network alignment problem aims at inferring the anchor links connecting the shared users across social networks, which are usually subject to the one-to-one cardinality constraint. Several existing social network alignment models have been proposed, many of which are based on the supervised learning setting. Given a set of labeled anchor links, a group of features can be extracted manually for the anchor links to build these models. Meanwhile, such methods may encounter great challenges in the application on real-world social network datasets, since manual feature extraction can be extremely expensive and tedious for the social networks involving heterogeneous information. In this paper, we propose to address the heterogeneous social network alignment problem with a deep learning model, namely DETA (Deep nETwork Alignment). Besides a small number of explicit features, DETA can automatically learn a set of latent features from the heterogeneous information. DETA models the anchor link one-to-one cardinality constraint as a mathematical constraint on the node degrees. Extensive experiments have been done on real-world aligned heterogeneous social network datasets, and the experimental results have demonstrated the effectiveness of the proposed model compared against the existing state-of-the-art baseline methods.

Community detection is an effective approach to unveil relationships among individuals in online social networks. In the literature, quite a few algorithms have been proposed to conduct community detection by exploiting the topology of social networks and the attributes of social actors. In practice, community detection is usually conducted by third parties, such as advertisement companies and hospitals, with access to social networks for different purposes, which can easily lead to a privacy breach. In this paper, we investigate community detection in social networks aiming to protect the privacy of both the network topology and the users' attributes. We show that with additional prior knowledge, community detection can be performed by querying the information of only a fraction of instead of the entire population. In particular, we first propose a new scheme called differentially private community detection (DPCD). DPCD detects communities in social networks via a probabilistic generative model, which can be decomposed into subproblems solved by individual users. The private social relationships and attributes of each user are protected by objective perturbation with differential privacy guarantees. Then, we propose a parsimonious node affiliation recovery (NAR) algorithm, which is also differentially private, to unveil the community affiliation information of the whole population based on that of the limited number of queried individuals by solving a sparse optimization problem. Through both theoretical analysis and experimental validation using synthetic and real-world social networks, we demonstrate that the proposed DPCD scheme detects social communities under the modest privacy budget. In addition, we show the effectiveness of NAR to perform community detection by querying a limited number of individuals in social networks.

In the world of e-marketing, new business models are introduced to accommodate changes caused by various factors, including the markets, the services, the customers, among others. One latest trend of e-marketing is social networking Web sites, many of which have attracted not only large number of users and visitors, but also business companies to place their online ads on the sites. As an important example of Web 2.0 applications, online social networks deserve comprehensive studying and analysis; they are not only employed as an effective vehicle of e-marketing, but may impact how future Web-based applications would be developed. In this chapter, we explore online social networking as a new trend of e-marketing, by conducting a comparative analysis of online social networking sites. We first discuss the various types of online social networks, based on the classification by Laudon & Traver (2008), and then analyze online social networks from a business strategy point of view, by discussing the primary revenue models for online social networking sites. The primary contribution of this chapter is a comparative analysis and discussions of representative online social networking sites and their respective revenue model(s). This chapter aims to provide the reader with a basic understanding of the emerging online social networking Web sites and their primary revenue models.

The increasing popularity and diversity of social media sites has encouraged more and more people to participate on multiple online social networks to enjoy their services. Each user may create a user identity, which can includes profile, content, or network information, to represent his or her unique public figure in every social network. Thus, a fundamental question arises -- can we link user identities across online social networks? User identity linkage across online social networks is an emerging task in social media and has attracted increasing attention in recent years. Advancements in user identity linkage could potentially impact various domains such as recommendation and link prediction. Due to the unique characteristics of social network data, this problem faces tremendous challenges. To tackle these challenges, recent approaches generally consist of (1) extracting features and (2) constructing predictive models from a variety of perspectives. In this paper, we review key achievements of user identity linkage across online social networks including stateof- the-art algorithms, evaluation metrics, and representative datasets. We also discuss related research areas, open problems, and future research directions for user identity linkage across online social networks.

Online social networking usage is growing rapidly, especially among at-risk populations, such as men who have sex with men (MSM). However, little research has studied the relationship between online social networking usage and sexual risk behaviors among at-risk populations. One hundred and eighteen Facebook-registered MSM (60.1% Latino, 28% African American; 11.9% other) were recruited from online (social networking websites and banner advertisements) and offline (local clinics, restaurants and organizations) venues frequented by minority MSM. Inclusion criteria required participants to be men who were 18 years of age or older, had had sex with a man in the past 12 months, were living in Los Angeles, and had a Facebook account. Participants completed an online survey on their social media usage and sexual risk behaviors. Results from a multivariable regression suggest that number of sexual partners met from online social networking technologies is associated with increased: 1) likelihood of having exchanged sex for food, drugs, or a place to stay within the past 3 months; 2) number of new partners within the past 3 months; 3) number of male sex partners within the past 3 months; and 4) frequency of engaging in oral sex within the past 3 months, controlling for age, race, education, and total number of sexual partners. Understanding the relationship between social media sex-seeking and sexual risk behaviors among at-risk populations will help inform population-focused HIV prevention and treatment interventions.

Union acceptable profit maximization in social networks

BackgroundThe dramatic growth of Web 2.0 technologies and online social networks offers immense potential for the delivery of health behavior change campaigns. However, it is currently unclear how online social networks may best be harnessed to achieve health behavior change.ObjectiveThe intent of the study was to systematically review the current level of evidence regarding the effectiveness of online social network health behavior interventions.MethodsEight databases (Scopus, CINAHL, Medline, ProQuest, EMBASE, PsycINFO, Cochrane, Web of Science and Communication & Mass Media Complete) were searched from 2000 to present using a comprehensive search strategy. Study eligibility criteria were based on the PICOS format, where “population” included child or adult populations, including healthy and disease populations; “intervention” involved behavior change interventions targeting key modifiable health behaviors (tobacco and alcohol consumption, dietary intake, physical activity, and sedentary behavior) delivered either wholly or in part using online social networks; “comparator” was either a control group or within subject in the case of pre-post study designs; “outcomes” included health behavior change and closely related variables (such as theorized mediators of health behavior change, eg, self-efficacy); and “study design” included experimental studies reported in full-length peer-reviewed sources. Reports of intervention effectiveness were summarized and effect sizes (Cohen’s d and 95% confidence intervals) were calculated wherever possible. Attrition (percentage of people who completed the study), engagement (actual usage), and fidelity (actual usage/intended usage) with the social networking component of the interventions were scrutinized.ResultsA total of 2040 studies were identified from the database searches following removal of duplicates, of which 10 met inclusion criteria. The studies involved a total of 113,988 participants (ranging from n=10 to n=107,907). Interventions included commercial online health social network websites (n=2), research health social network websites (n=3), and multi-component interventions delivered in part via pre-existing popular online social network websites (Facebook n=4 and Twitter n=1). Nine of the 10 included studies reported significant improvements in some aspect of health behavior change or outcomes related to behavior change. Effect sizes for behavior change ranged widely from −0.05 (95% CI 0.45-0.35) to 0.84 (95% CI 0.49-1.19), but in general were small in magnitude and statistically non-significant. Participant attrition ranged from 0-84%. Engagement and fidelity were relatively low, with most studies achieving 5-15% fidelity (with one exception, which achieved 105% fidelity).ConclusionsTo date there is very modest evidence that interventions incorporating online social networks may be effective; however, this field of research is in its infancy. Further research is needed to determine how to maximize retention and engagement, whether behavior change can be sustained in the longer term, and to determine how to exploit online social networks to achieve mass dissemination. Specific recommendations for future research are provided.

A deep dive into user display names across social networks

In the last decade, online social networking websites have become immensely popular worldwide, with the number of users growing exponentially. With a social network account, users can post an array of personal information and link their profile to those of other users, forming virtual communities. In the last few years, security and protection of private information posted on social networking sites has piqued the interest of researchers. Generally, the focus of this research has been limited to describing the type of information posted by users, with few studies examining the relationship between users' adoption of security measures for their social networking sites and negative online experiences, such as online interpersonal victimization. Our examination of the relationship between users' online social network activity, online social network security, and online interpersonal victimization found that users who engaged in risky online behaviors, such as opening numerous social network accounts and adding strangers as friends, were more likely to be victimized online.

Online social networks (OSNs) have gained soaring popularity and are among the most popular sites on the Web. With OSNs, users around the world establish and strengthen connections by sharing thoughts, activities, photos, locations, and other personal information. However, the immense popularity of OSNs also raises significant security and privacy concerns. Storing millions of users' private information and their social connections, OSNs are susceptible to becoming the target of various attacks. In addition, user privacy will be compromised if the private data collected by OSNs are abused, inadvertently leaked, or under the control of adversaries. as a result, the tension between the value of joining OSNs and the security and privacy risks is rising.;To make OSNs more secure and privacy-preserving, our work follow a bottom-up approach. OSNs are composed of three components, the infrastructure layer, the function layer, and the user data stored on OSNs. For each component of OSNs, in this dissertation, we analyze and address a representative security/privacy issue. Starting from the infrastructure layer of OSNs, we first consider how to improve the reliability of OSN infrastructures, and we propose Fast Mencius, a crash-fault tolerant state machine replication protocol that has low latency and high throughput in wide-area networks. For the function layer of OSNs, we investigate how to prevent the functioning of OSNs from being disturbed by adversaries, and we propose SybilDefender, a centralized sybil defense scheme that can effectively detect sybil nodes by analyzing social network topologies. Finally, we study how to protect user privacy on OSNs, and we propose two schemes. MobiShare is a privacy-preserving location-sharing scheme designed for location-based OSNs (LBSNs), which supports sharing locations between both friends and strangers. LBSNSim is a trace-driven LBSN model that can generate synthetic LBSN datasets used in place of real datasets. Combining our work contributes to improving security and privacy in OSNs.

As per recent progress, online social network (OSN) users have grown tremendously worldwide, especially in the wake of the COVID-19 pandemic. Today, OSNs have become a core part of many people’s daily lifestyles. Therefore, increasing dependency on OSNs encourages privacy requirements to protect users from malicious sources. OSNs contain sensitive information about each end user that intruders may try to leak for commercial or non-commercial purposes. Therefore, ensuring different levels of privacy is a vital requirement for OSNs. Various privacy preservation methods have been introduced recently at the user and network levels, but ensuring k-anonymity and higher privacy model requirements such as l-diversity and t-closeness in OSNs is still a research challenge. This study proposes a novel method that effectively anonymizes OSNs using multiple-graph-properties-based clustering. The clustering method introduces the goal of achieving privacy of edge, node, and user attributes in the OSN graph. This clustering approach proposes to ensure k-anonymity, l-diversity, and t-closeness in each cluster of the proposed model. We first design the data normalization algorithm to preprocess and enhance the quality of raw OSN data. Then, we divide the OSN data into different clusters using multiple graph properties to satisfy the k-anonymization. Furthermore, the clusters ensure improved k-anonymization by a novel one-pass anonymization algorithm to address l-diversity and t-closeness privacy requirements. We evaluate the performance of the proposed method with state-of-the-art methods using a “Yelp real-world dataset”. The proposed method ensures high-level privacy preservation compared to state-of-the-art methods using privacy metrics such as anonymization degree, information loss, and execution time.

Online social networks are vital environments for information sharing and user interactivity. To help users of online social services to build, expand, and maintain their friend networks or webs of trust, trust management systems have been deployed and trust inference (or more generally, friend recommendation) techniques have been studied in many online social networks. However, there are some challenging issues obstructing the real-world trust inference tasks. Using only explicit yet sparse trust relationships to predict user trust is inefficient in large online social networks. In the age of privacy-respecting Internet, certain types of user data may be unavailable, and thus existing models for trust inference may be less accurate or even defunct. Although some less interpretable models may achieve better performance in trust prediction, the interpretability of the models may prevent them from being adopted or improved for making relevant informed decisions. To tackle these problems, we propose a probabilistic graphical model for trust inference in online social networks in this paper. The proposed model is built upon the skeleton of explicit trust relationships (the web of trust) and embeds various types of available user data as comprehensively-designed trust-aware features. A message passing algorithm, loop belief propagation, is applied to the model inference, which greatly improves the interpretability of the proposed model. The performance of the proposed model is demonstrated by experiments on a real-world online social network dataset. Experimental results show the proposed model achieves acceptable accuracy with both fully and partially available data. Comparison experiments were conducted, and the results show the proposed model’s promise for trust inference in some circumstances.

With the increasing number of online social networking site users, there is increasing cyber attacks on online social networks. Strong connectivity among users of the social networking site makes the information of interest rapidly spread. If the worm disguised as information to attract users' interest spreads on online social networking sites, it can cause great damage. Therefore, a fast patch propagation schemes need to be able to inhibit the activity of the worm. In this paper, we propose a fast patch propagation scheme based on genetic algorithms for online social networks.

Using communities against deception in online social networks

Online social media network has become part of human life by transforming the way users create new social relations or relate with family and friends. Online social network (OSN) has drastically increased the rate at which people interact with each other by simplifying the means of communication. However, privacy is raising a serious concern. All user generated data within the OSN system need to be protected against unauthorized friends or hackers or even against the provider of OSN. Many research works are going on to encounter the privacy issues in OSN. This paper analyses the limitations of the recent work being done in this field and proposes an efficient abstract solution to them.