Differential privacy: a solution to privacy issue in social networks

Preeti Malik Graphic Era (Deemed to be) University, Dehradun, INDIA Varsha Mittal Graphic Era Deemed to be University, Dehradun, India Mohit Mittal INRIA Labs, France Kamika Chaudhary MB Govt. PG College, Haldwani, India

Abstract

The privacy of social network data is becoming increasingly important, threatening to limit access to this lucrative data source. The topological structure of social networks can provide useful information for income production and social science research, but it is challenging to ensure that this analysis does not breach individual privacy. Differential privacy is a prominent privacy paradigm in data mining over tabular data that employs noise to disguise individuals' contributions to aggregate findings and provides a very exceptional analytical guarantee that individuals' existence in the data-set is hidden. Because social network analysis has multiple applications, it opens up a new field for differential privacy applications. This article provides a thorough examination of the fundamental principles of differential privacy and their applications in computing.

Keywords: Differential Privacy, global sensitivity, smooth sensitivity, degree distribution,

1 Social media and its popularity

The growth of social media began in 1996 with the debut of the networking site Bolt (now closed).1 Soon later, in 1997, Six Degrees was launched, allowing users to add friends and establish profiles. Following that, programs such as AOL Instant Messenger, Live Journal, and Friendster were created, all of which helped pave the way for Facebook to launch in 2004. Every day, more people are using social media. The number of active social media users worldwide reached 4.48 billion in 2021, an increase of 13.13% from 3.69 billion in 2020. In 2015, there were just 2.07 billion users, suggesting a 115.59% increase in just 6 years.

1.1 Pandemic marketing update

In July 2020, DataReportal produced a unique report that examines changes in social media activity at the commencement of the COVID-19 lockdown period, in addition to usual enquiries. The amount of Internet and digital activities has increased dramatically (see Figure 1).

Figure 1: Impact of COVID-19 on online activities.2

1.2 Pros and cons of using social media

1.2.1 Pros

There are several more advantages of using social media:

• Digital media knowledge: It allows your kid to explore and experiment on social media. It also aids them to get the material and proficiencies they need to adore online events while escaping from online risks.

• Cooperative learning: Your kid may exchange educational information on social media.

• Creativity: Your youngsters may express themselves through their profile pages, images, and videos.

• Mental health and well-being: Interacting with people and friends on social media provides an emotion of belonging and connection in your kid.

1.2.2 Cons

Social media may sometimes be dangerous. The dangers for your kids include:

• Uncovering aggressive or distressing information, like harsh, offensive, violent, or sexual remarks or snaps.

• Sharing wrong content, for instance, snaps or videos that are uncomfortable or suggestive.

• Sharing personal information on social media with strangers, for example, contact number, birth date, or addresses. Privacy settings can limit who can view information about your kids, such as their name, age, and where they reside. One can misuse this information.

• One can become victim of cyberbullying.

2 Social network analysis

Disease transmission, emotional contagion, and professional mobility are all examples of critical societal concerns that may be discovered via social network research [1, 5]. Social networks are designed to distribute data without revealing personal information due to the requirement for scientific study and data exchange. The original data can be disturbed or encrypted, or anonymous processing can be performed before releasing the data [2, 3, 4].

The phrase "privacy" is loaded, since it means different meaning for different people. Edge weights in social networks may indicate the frequency of contact, the cost of economic exchange, the closeness of a connection, and other factors that are linked to sensitive data. An intelligence network is a good example, where edge weights represent the frequency of communication between two organizations. Excessive communication might indicate an issue. A commercial trade network is another illustration, where edge weights represent the price of a transaction between two businesses. Due to the severe rivalry, most managers would be hesitant to give a business secret to their competitors. Our objective is to prevent edge weight leaking in social networks while retaining as much usefulness as feasible.

Dalenius [6] first diagnosed privacy protection issue in the late 1970s. According to Dalenius, privacy protection aimed at preventing any user either legitimate or spurious, from accessing original data of any individual while accessing the database. A number of solutions have been proposed by researchers which are based on this idea, including k-anonymity [7], l-diversity [8], t-closeness [9], and (a, k)-anonymity [10]. Though all these models protect against a certain form of assault and are unable to fight against newly invented attacks, the security of the model is based on the hypothesis of some specific background information of an attacker, which is a primary source of this flaw. Nonetheless, enumerating all conceivable sorts of background information of attacker may have been very hard. As a result, a model that preserves privacy while ignoring background knowledge is very desired.

3 Privacy breaches in social networks

Defining the term privacy breach is crucial [11]. When a bit of delicate information of an individual is given to an enemy or to someone having the objective of damaging privacy, it is called a privacy breach. Identity disclosure and attribute disclosure are the two forms of privacy breaches that have traditionally been researched. In the framework of social networks, we explore these two forms. We also discuss two other forms of network data disclosures: social link and affiliation link disclosure.

3.1 Identity disclosure

When a challenger is capable of discovering the mapping from a social network profile p to a particular real-world entity i, identity exposure happens. Let us analyze three issues about i's identity in which an opponent would be interested before we can establish a formal definition of identity disclosure. These definitions (see Table 1) have been taken from Zheleva and Getoor [11].

Table 1:Query definitions.

Definition 1 (Mapping query). In a set of individual profiles (P) in a social network G, find which profile p maps to a particular individual i. Return p. Definition 2 (Existence query). For a particular individual i, find if this individual has a profile p in the network G. Return true or false. Definition 3 (Co-reference resolution query). For two individual profiles pi and pj, find if they refer to the same individual...