Structural vulnerabilities and Link Privacy in Social Networks
Introduction/Background:
In social networks, a link represents a relationship between two nodes in the network. These links can represent email conversations, web surfing, co-purchases of two or more products (e.g. Amazon), friendships (e.g. Facebook), followers (e.g. Twitter), etc. Often times these relationships are sensitive and/or confidential in nature [ying-wu] and the users are operating under the assumption that their private relationships will not be disclosed.
In recent years the amount of data accumulated from social networks has become very large, and there is a lot of valuable information to gain from analyzing and applying data mining to social network data.
…show more content…
Results:
Neighborhood Randomization Using Sub-Graph Perturbation
In order for people to mine valuable data from social network graphs they must first be given information about the network. Even without explicit information about the nodes, an attacker may use structural information about the nodes and graph itself (e.g. node degree) to identify who the individuals are that the nodes represent. Simple graph-wise randomization addresses this problem by deleting k randomly chosen edges and replacing them with k randomly chosen edges, however a problem arises since data-miners depend on these structural attributes to properly analyze the social network. Fard and Wang [fard-wang] propose a structure-aware algorithm for the randomization of social network edges as well as a formal definition of “link privacy“ with respect to a probabilistic threshold. Their motivation is to help conceal sensitive links by using randomization techniques, without disturbing the actual structure of the graph, which is achieved through local neighborhood perturbation. This is needed so that graphs can be analyzed without the link structure being left entirely vulnerable to attackers. The goal of their algorithm is to make it so that an adversary cannot know if a link in the original graph exists from having a link in the new graph.
Problem definition: "Given a
It has been the ultimate norm to most of us to publicize almost everything we do in our personal lives on the Internet. With the help of various social networking sites (SNS), such as Facebook, Twitter, and Instagram, we have been able to effortlessly keep in touch with our dear relatives and friends. Unfortunately, many of us may have forgotten the importance of sheer privacy and that whatever we post online will be forever public, even after its deletion, for everyone to see somewhere in the depths of the Internet. We have excessively exposed an enormous chunk of our personal data online including our relationships. It may be easy to link with someone, but severing ties with them, especially when the information has long been published on
In the last decade, social networks like Facebook [1] have emerged as popular medium of social interaction and information dissemination. From a social web data mining perspective, Facebook stores a wealth of data about people and their interests. As more and more users are creating their own content on Facebook, there is a growing interest to mine this data for use in personalized information access services, recommender systems, tailored advertisements, and other applications that can benefit from personalization.
Social Network Theory, as we know it today, refers to the structural analysis of social relationships between people, based on systematic empirical data (Marsden, 1990). These relations are commonly displayed by nodes and ties, where nodes represent the individuals within the network and ties the connection between them, showing the degree of interdependency such as close friendship or loose business contact (illustrated in figure 1). Generally, these connections can be divided into strong and weak ties in order to denote the respective closeness of the individuals’ relationships.
Social network analysis is used to study the pattern of communication and relationships among the members of a social network. The interaction on the social network is assumed to be reflective of how the individual interacts and is an insight into the members’ behavior patterns (Costa, 2012). The graph theory has been used effectively to understand a variety of unrelated problems from organizational behavior to spread of infectious diseases. In social network research, various algorithms have been designed in conjunction with concepts from relational learning, web mining and inductive logic to perform predictive as well as descriptive analysis. As described in Karamon (2008) some of
Using data mining techniques, such as graph mining and social network analysis on regional data sources could contribute great insights and improve operations. Social networking analysis is the study and analysis of networks involving social interaction. Types of
Our research rests upon two pillars. The first one is related to how we conceptualize the privacy concept and the other is the model proposed by Miltgen and Reyrat-Guillard (2014). In our research, we focus on information privacy, which can be categorized as a sub-construct of the general privacy construct (Smith et. al., 2011). More specifically, we define information privacy as factual privacy, which deals with “issues about the access, collection, dissemination and use of personal factual information” within the context of online social network sites (Zhang et. al., 2011).
Third, conceiving anonymization strategies for informal organization information is much harder than that for social information. Anonymizing a gathering of tuples in a social table does not influence other tuples. On the other hand, while altering a system, changing one vertex or edge may influence whatever remains of the system. Along these lines, ``divide-and-vanquish'' systems, which are generally connected to social information, can't be connected to network information. To manage above difficulties, numerous methodologies have been proposed. As indicated by [15], anonymization
A social network is a set of people (or organizations or other social entities) connected by a set of social relationships such as friendship, co-working or information exchange. Social networks are connected through various social familiarities ranging from casual acquaintance to close familiar bonds. Social network analysis provides both a visual and a mathematical analysis of relationships. Social network analysis (SNA) is a quantitative analysis of relationships between individuals or organizations. By quantifying such social structures it is possible to identify most important actors, group formations or equivalent roles of actors within a social network. This paper presents various properties or analysis measures for social
The Facebook dataset [12] was present in the form of an edgelist. The dataset was read and its corresponding graph was generated, which serves as the original network representation. The random graphs – Erdos–Renyi (ER) random
The widespread use of social media sites such as Facebook, Instagram, Twitter and LinkedIn have been generating large amounts of data which are interchangeably known as social data. This data is usually user-generated, inconsistent, and incomplete and have geo information. This data can be used by researchers or companies to extract meaningful information about human behavior, product consumption, trends in markets, etc. Extraction and study of this data for useful purposes is called data mining of social media. Since the number of techniques used to mine data are increasing rapidly with time, it has become imperative to represent the data in a form which can be easily understood. Data mining may consist of the following stages. First, data is extracted,
Online Social Network (OSN) sites act as a medium to spread their own views, activities and their thoughts to some camaraderie. The OSNs contents and images are spread on the web isn’t determined by a human decision. At present, there is a risk of information leakage in many social networking sites. There is no mechanism used
Abstract—In recent years social network has became pervasive all over the world, and smartphone plays an important role for individuals communication. Online social networking is a pervasive communication platform where users with smart- phones or website can search over the Internet and query neighboring peers to obtain the information they desired. We do research on main techniques that can protect user’s privacy and build robust and trustful application. Moreover, we deeply discuss the issues of online social network and the problem we face to. Future trends has also been discussed, we try to figure out a good way to build a trustful and solid social network.
Social networks are popular infrastructures for communication, relations, and information sharing on the internet, popularly known as social network such as My Space Twitter and Facebook supply communication, storage and social applications for hundreds of millions users. Users set up social links to friends and influence their social links to distribute content, organise measures for particular users or share resources, therefore, these social networks provide proposal for organising event users to users communication, and also between the internet’s most accepted destinations. Currently work has been seen as a form of group of socially improved applications that influence the correlation from social networks to improve security and performance of online social network applications as well as spam email improvement Garriss, (2006) On one hand, significant interactive with friends relationship are essential to developed confidence and trustworthiness in the system.
Online Social Networks (OSNs), a very popular application on the Internet, have attracted almost one billion users, many of whom have incorporated these applications into their daily practices [Deep Nishar. April 18, 2014, Twitter Inc. June 2014, Socialbakers. , Google Official Blog. April 11, 2012] . Nowadays, there are hundreds of OSN sites which facilitate and enable the users to interact and collaborate with each other in a virtual community. The rapid rise of a large variety of OSN sites, with the massive amount of available information, obviously raises new, serious concerns about the security and privacy of their users and requires insights into security and privacy issues. Researchers from different computer science disciplines have investigated some of the privacy and security problems which arise in OSNs, from different viewpoints (e.g., [Raad and Chbeir. 2013, Pesce, et al. 2012, Hu and Ahn. 2011, Gurses and Diaz. 2013, Hongyu Gao, et al. 2011, Mahmood. 2013] ). As a shared platform, the lack of collaborative policy for access control and authorization management has become one of the most important and crucial issues in OSNs. Currently, OSNs have limited access control where the privacy settings of shared content is solely defined and regulated by the uploader of the shard content, regardless of other involved users. Hence, because of the limited and poor access control mechanisms for shared data in OSNs, the concerns of information
Know a days social network is used by number of people in the universe. Increase in the use of social networks in mobile causing the security and privacy problems. Social information is related to their address, locations, relationships in real-time according to their activities. Some of the systems such as Serendipity and WhozThat which provides security in social networks.