社交网络中用户隐私推理与保护研究综述

如今微博和Twitter等社交网络平台被广泛地用于交流、创建在线社区并进行社交活动。用户所发布的内容可以被推理出大量隐私信息,这导致社交网络中针对用户的隐私推理技术的兴起。利用用户的文本内容及在线行为等知识可以对用户进行推理攻击,社交关系推理和属性推理是对社交网络用户隐私的两种基本攻击。针对推理攻击保护机制和方法的研究也在日益增加,对隐私推理和保护技术相关的研究和文献进行了分类并总结,最后进行了探讨和展望。     

IncogniSense: An anonymity-preserving reputation framework for participatory sensing applications

Reputation systems are fundamental for assessing the quality of user contributions in participatory sensing. However, naively associating reputation scores to contributions allows adversaries to establish links between multiple contributions and thus de-anonymize users. We present the IncogniSense framework as a panacea to these privacy threats. IncogniSense utilizes periodic pseudonyms generated using blind signature and relies on reputation transfer between these pseudonyms. Simulations are used to analyze various reputation cloaking schemes that address the inherent trade-off between anonymity protection and loss in reputation. Our threat analysis confirms the robustness of IncogniSense and a prototype demonstrates that associated overheads are minimal.     

The effect of role conflict on self‐disclosure in social network sites: An integrated perspective of boundary regulation and dual process model

As people increasingly integrate social network sites (SNSs) into their daily lives, their social connections on SNSs become more diverse, including their friends, co‐workers, and relatives. In such a context, users may receive different role expectations from their various social circles and experience role conflict in their usage of SNSs. Furthermore, previous literature suggests that people may not always make privacy‐related decisions through effortful and deliberate information processing. Drawing on the perspective of boundary regulation and dual process theories, this study clarifies the consequences of role conflict on SNSs. A theoretical model is developed to examine the effect of role conflict regarding privacy risk and perceived control, which, in turn, impact self‐disclosure behaviour, as well as how this process is moderated by high‐ versus low‐effort processing. The results of 4 experiments provide strong support for our model. The implications for theory and practice are discussed.     

社交网Twitter平台的人物关系网社区发现

社交网络服务（social networking service,SNS）已融入到大众生活中。人们将自己的信息上传到网络中,并通过社交网站管理自己的社交圈子,由此造成大量的个人信息在社交网络上被公开。文章基于Twitter平台,设计实现了Twitter用户关系网的社区发现。通过实时采集Twitter用户信息,重建人物关系网,改进Newman快速算法划分社区发现人物关系网。文章通过可视化的界面呈现用户的社区关系,提供用户网络行为,为决策者的舆情监控或个性推荐提供了参考凭据。     

社会网络的隐私保护研究综述*

某些网站将匿名处理后的社会网络数据公开发布,或者提供给科研机构、大学或者其他组织和个人使用,而这些数据往往侵犯了用户的隐私,但有关社会网络中个人信息安全和隐私保护的研究却处于起步阶段。综述了当前在线社会网络的研究成果,主要就社会网络及其隐私漏洞、信息泄露、再识别攻击、聚集攻击、推理攻击等进行了分析,并对今后的发展提出了预测,为社会网络的科研指出了可行的研究方向。     

Structure of Ego‐Alter Relationships of Politicians in Twitter

We analyze the ego‐alter Twitter networks of 300 Italian MPs and 18 European leaders, and of about 14,000 generic users. We find structural properties typical of social environments, meaning that Twitter activity is controlled by constraints that are similar to those shaping conventional social relationships. However, the evolution of ego‐alter ties is very dynamic, which suggests that they are not entirely used for social interaction, but for public signaling and self‐promotion. From this standpoint, the behavior of EU leaders is much more evident, while Italian MPs are in between them and generic users. We find that politicians – more than generic users – create relationships as a side effect of tweeting on discussion topics, rather than by contacting specific alters.     

Social media and our misconceptions of the realities

This article will review the current laws of the expectations of privacy under the two-pronged Katz test, and more specifically other cases that the courts have tried to interpret the test as applicable to social media such as Facebook, Instagram, Twitter, and email. Since there seems to be ‘no light at the end of the tunnel’ with any uniform decision within the courts on the Fourth Amendment protections against search and seizure when there is a reasonable expectation of privacy with social media. This reasonable expectation standard is developed by the users themselves who will allow their rights to be infringed upon for the approval to access sites like Facebook, Twitter, and Instagram.     

位置服务中基于二分图的身份推理攻击算法

针对位置服务中的身份隐私泄露问题,提出了一种基于二分图的身份推理攻击算法。其基本思想是构建移动用户真实身份和假名间的有权二分图,运用Kuhn-Munkres算法找到其最佳完美匹配,确定用户的真实身份完成攻击。通过实验验证了该算法的有效性,并分析了隐私保护机制、位置服务隐私泄露率和假名生存期等因素对算法的影响。     

基于节点分割的社交网络属性隐私保护

现有研究表明,社交网络中用户的社交结构信息和非敏感属性信息均会增加用户隐私属性泄露的风险.针对当前社交网络隐私属性匿名算法中存在的缺乏合理模型、属性分布特征扰动大、忽视社交结构和非敏感属性对敏感属性分布的影响等弱点,提出一种基于节点分割的隐私属性匿名算法.该算法通过分割节点的属性连接和社交连接,提高了节点的匿名性,降低了用户隐私属性泄露的风险.此外,量化了社交结构信息对属性分布的影响,根据属性相关程度进行节点的属性分割,能够很好地保持属性分布特征,保证数据可用性.实验结果表明,该算法能够在保证数据可用性的同时,有效抵抗隐私属性泄露.     

A study of social networking site use from a three-pronged security and privacy threat assessment perspective

This study focuses on how social networking site (SNS) users’ perceived risk is affected by perceptions of the duality of potentially harmful events (cyber-attacks and privacy breaches) on SNS and explores the potential threat sources that influence such perceptions. Further, it examines the effects of perceived risk on SNS members’ site use and the underlying mechanisms through which these effects are cast. The study finds that users’ perceived SNS risk affects site use behavior through attitude as an important mediator and users’ concern regarding cyber-attacks (security) casts a greater impact than the concern regarding privacy breaches in shaping perceived SNS risk.     

Using Estimated Personality of Social Network Members for Finding Influential Nodes in Viral Marketing

Sending promotional messages to a few numbers of users in a social network can propagate a product through word of mouth. However, choosing users that receive promotional messages, in order to maximize propagation, is a considerable issue. These recipients are named “influential nodes.” To recognize influential nodes, according to the literature, criteria such as the relationships of network members or information shared by each member on a social network have been used. One of the effective factors in diffusion of messages is the personality characteristics of members. As far as we know, although this issue is considerable, so far it has not been applied in the previous studies. In this article, using the graph structure of social networks, two personality characteristics, openness and extroversion, are estimated for network members. Next, these two estimated characteristics together with other characteristics of social networks, are considered as the criteria of choosing influential nodes. To implement this process, the real coded genetic algorithm is used. The proposed method has been evaluated on a dataset including 1000 members of Twitter. Our results indicate that using the proposed method, compared with simple heuristic methods, can improve performance up to 37%.     

Privacy-Aware Framework for Matching Online Social Identities in Multiple Social Networking Services

With various emerging Social Networking Services (SNS), it is possible for users to join multiple SNS for social relationships with other users and to collect a large amount of information (e.g., statuses on Facebook and tweets on Twitter). However, these users have been facing difficulties in managing all the data collected from the multiple SNS. It is important to match social identities from the multiple SNS. In this study, we propose a privacy-aware framework for a social identity matching (SIM) method across these multiple SNS. It means that the proposed approach can protect user privacy, because only the public information (e.g., username and the social relationships of the users) is employed to find the best matches between social identities. As a result, we have shown by evaluation that the F-measure of the proposed SIM method is about 60%.     

College students and risk-taking behaviour on Twitter versus Facebook

The propensity of college students to post content that they know may be unacceptable to future employers or other authority figures has been well established. Yet research on this topic has tended to focus exclusively on Facebook, which is problematic for two reasons. First, many young social media users are shifting away from Facebook and towards Twitter and other services. Second, college students have changed their use of social media over time and may now be more cautious about what they post on Facebook. To address this issue, a survey-based field study was conducted to compare student comfort levels with authority figures viewing their Facebook and Twitter accounts. Specifically, undergraduate business students attending a large university in the midwest of the USA were surveyed about their Facebook and Twitter accounts. Findings indicate that college students are markedly less comfortable with authority figures viewing their Twitter accounts. Paradoxically, a great majority of the study respondents were found to have public Twitter accounts, while only a very small minority have public Facebook accounts. This finding suggests that students perceive less risk on Twitter versus Facebook or that they are writing to different imagined audiences on the two platforms. Implications include the need for further inquiry and an awareness of educators and human resources professionals about students’ current social media practices.     

PerContRep: a practical reputation system for pervasive content services

Social network has extended its popularity from the Internet to mobile domain. Personal mobile devices can be self-organized and communicate with each other for instant social activities at any time and in any places to achieve pervasive social networking (PSN). In such a network, various content information flows. To which extent should mobile users trust it, whilst user privacy can also be preserved? Existing work has not yet seriously considered trust and reputation management, although trust plays an important role in PSN. In this paper, we propose PerContRep, a practical reputation system for pervasive content services that can assist trustworthy content selection and consumption in a pervasive manner. We develop a hybrid trust and reputation management model to evaluate node recommendation trust and content reputation in the context of frequent change of node pseudonyms. Simulations show the advantages of PerContRep in assisting user decisions and its effectiveness with regard to unfair rating attack, collaborative unfair rating attack, on-off attack and conflict behavior attack. A prototype system achieves positive user feedback on its usability and social acceptance.     

Relationship Identification Across Heterogeneous Online Social Networks

In the era of the social web, many people manage their social relationships through various online social networking services. It has been found that identifying the types of social relationships among users in online social networks facilitates the marketing of products via electronic “word of mouth.” However, it is a great challenge to identify the types of social relationships, given very limited information in a social network. In this article, we study how to identify the types of relationships across multiple heterogeneous social networks and examine if combining certain information from different social networks can help improve the identification accuracy. The main contribution of our research is that we develop a novel decision tree initiated random walk model, which takes into account both global network structure and local user behavior to bootstrap the performance of relationship identification. Experiments conducted based on two real‐world social networks, Sina Weibo and Jiepang, demonstrate that the proposed model achieves an average accuracy of 92.0%, significantly outperforming other baseline methods. Our experiments also confirm the effectiveness of combining information from multiple social networks. Moreover, our results reveal that human mobility features indicating location categories, coincidence, and check‐in patterns are among the most discriminative features for relationship identification.     

基于社交圈层和注意力机制的信息热度预测

社交网络现已成为现实世界中信息传播与扩散的主要媒介,对其中的热点信息进行建模和预测有着广泛的应用场景和商业价值,比如进行信息传播挖掘、广告推荐和用户行为分析等.目前的相关研究主要利用特征和时间序列进行建模,但是并没有考虑到社交网络中用户的社交圈层对于信息传播的作用.本文提出了一种基于社交圈层和注意力机制的热度预测模型SCAP(Social Circle and Attention based Popularity Prediction),首先对社交圈层进行定义,通过自动编码器提取用户历史文本序列的特征,对不同用户的社交圈层进行聚类划分,得到社交圈层特征.进而对于一条新发布的文本信息,通过长短期记忆网络与嵌入层提取其文本特征、用户特征和时序特征,并基于注意力机制,捕获到不同社交圈层对于该文本信息的影响程度,得到社交圈层注意力特征.最后将文本特征、用户特征、时序特征和社交圈层注意力特征进行特征融合,并通过两个全连接层进行建模学习,对社交信息的热度进行预测.在推特、微博和豆瓣等四个数据集上的实验结果表明,SCAP模型的预测表现相比于多个对比模型总体呈优,在不同数据集上均方误差(MSE)分别降低了0.017,0.022,0.021和0.031,F1分数分别提升0.034,0.021,0.034和0.025,能够较为准确地预测社交信息的热度.本文同时探究了不同实验参数对于模型的影响效果,如用户历史文本序列的数量、社交圈层的数量和时间序列的长度,最后验证了模型输入的各个特征和注意力机制的引入对于模型预测性能提升的有效性,在推特数据集中,引入社交圈层和注意力机制,模型的MSE指标分别降低了0.065和0.019.     

Long Chains or Stable Communities? The Role of Emotional Stability in Twitter Conversations

In this article, we address the issue of how emotional stability affects social relationships in Twitter. In particular, we focus our study on users’ communicative interactions, identified by the symbol “@.” We collected a corpus of about 200,000 Twitter posts, and we annotated it with our personality recognition system. This system exploits linguistic features, such as punctuation and emoticons, and statistical features, such as follower count and retweeted posts. We tested the system on a data set annotated with personality models produced by human subjects and against a software for the analysis of Twitter data. Social network analysis shows that, whereas secure users have more mutual connections, neurotic users post more than secure ones and have the tendency to build longer chains of interacting users. Clustering coefficient analysis reveals that, whereas secure users tend to build stronger networks, neurotic users have difficulty in belonging to a stable community; hence, they seek for new contacts in online social networks.     

支持隐私保护的社交网络信息传播方法

社交网络影响力传播重点关注如何使用少量的种子集合在社交网络中产生尽可能高的影响力,并将转发作为信息传播的唯一方式,忽略了其他传播方式,例如用户可通过发布一条与所见信息内容相似的信息来进行传播,这种传播方式(称为转述)因为难以追踪,所以存在隐私泄漏的风险。针对上述问题,定义了一种支持转述关系的社交网络信息传播模型,提出了一种支持用户隐私保护的信息传播方法 LocalGreedy,确保用户发送的信息不泄漏到指定黑名单的同时,最大化传播产生的影响力,平衡了隐私保护和信息传播的矛盾。针对种子集合选取的枚举问题,提出了支持隐私保护的递增策略构造种子集合,减少时间开销;给出了计算节点的局部影响子图方法,快速估计种子集合传播产生的影响力;为确保种子集合满足隐私保护约束限制,提出了推导节点泄漏态概率上限的方法,避免使用蒙特卡洛方法产生的时间开销。使用爬取的新浪微博数据集进行实验验证和实例分析,结果表明了所提方法的有效性。     

Understanding online consumers’ purchase intentions: a contribution from social network theory

There is a dynamic and interconnected international setting shaped by the power of the Internet and social media. To gain more consumers, understand their behaviours and needs, and maintain closest relationships with them, businesses should understand how consumers behave in social media and how they vary in their purchase intentions. In the scope of the study, we integrate the social network theory and the theory of planned behaviour to analyse online consumers’ purchase intentions and to investigate their structural positions by analysing their friendships in social networks. We target Twitter users to conduct analysis due to Twitter's popularity in use, market penetration, and opportunity to work with open-source data. This study contributes to a better theoretical understanding of online consumers’ purchase intentions by integrating multiple theoretical perspectives. It expands the literature by considering both online consumers’ friendship network in Twitter and their individual online purchasing intentions. The study also guides e-marketers to design proper strategies for potential and current consumers and target the right sets of people in the social networks.     

基于社交网络好友攻击的位置隐私保护模型

随着无线网络的发展,移动社交网络用户发布其所在的地理位置信息时,如果包含敏感地理位置会导致用户隐私受到攻击。现有的位置隐私保护方法都是对用户发布的位置进行泛化处理,以牺牲用户的服务质量为代价,且大部分都是将攻击者定位在LBS服务商,没有考虑到统一对社交网络中的好友根据其可靠程度的不同提供不同准确度的地理位置信息。针对此问题,提出了基于社交网络好友亲密度分级的隐私保护模型L-intimacy,用来防止好友攻击者的攻击。理论分析和实验结果表明,与加入到Latitude服务的Google Maps相比,该方法既能保护移动社交网络用户的相关隐私,同时又具有较小的信息损失度。