Abstract
This paper discusses the issue of privacy in social networks with respect to advances in machine learning. It shows how machine learning protocols have been developed both to enhance and secure privacy as well as to invade privacy and collect, analyze, predict data based on users’ information and experience online. The conflict between these two directions in machine learning is likely to lead to a system wherein machine learning algorithms are actively engaged in the subversion of one another, with one attempting to conceal data and the other attempting to uncover it. This paper concludes with recommendations for social networks and the issue of privacy regarding machine learning.

Introduction

Social networks have allowed an ocean of personal data to form that is now sitting there waiting for machine learning algorithms to collect it, analyze it, and recognize individuals on social media (Oh, Benenson, Fritz & Schiele, 2016). Machine learning algorithms are thus being used more and more in social networks to collect data on users and to assess their browsing and personal information—and in doing so they could soon be predicting someone’s recreational activities or political affiliation through a simple analysis of an individual’s social media use, such as posts on Twitter or the friends one has on Facebook (Lindsey, 2019). As a result, the privacy of individual social media users may be in jeopardy. This paper will review the findings of the related literature on this subject and discuss them and the recommendations for addressing this issue in the future.

Review of Literature

Privacy and information sharing may seem like two diametrically opposed concepts in the context of social media, and to a high degree they are. Mobile devices allow users to set information sharing settings that allow algorithms on other applications to identify a person’s location, habits, and view other information to personalize ads and so on. The information that is available for viewing by machine learning programs is enormous, and many users do not even realize it. Machine learning algorithms often know more about a user’s habits and choices than the user does. Bilogrevic et al. (2016) point out that “by analyzing people’s sharing behaviors in different contexts, it is shown in these works that it is possible to determine the features that most influence users’ sharing decisions, such as the identity of the person that is requesting the information and the current location” (p. 126). The problem is that users do not know how to articulate their own personal information settings desires, nor are these desires static. For that reasons Bilogrevic et al. (2016) created a program that uses machine learning AI to automatically set those settings based on user interaction on the Web, reducing the user’s worry about when it is appropriate to share information and when it is not. The program developed by Bilogrevic et al. (2016) will do it for them.

Such a program is one example of the way privacy concerns and machine learning advancements are meeting in the realm of social media. One reason it is needed is, as Oh et al. (2016) show, there is no such thing as privacy on the Internet, and AI is being developed to collect as much data on what is out there as possible. The implications for one’s privacy are enormous, especially as more and more people put their entire lives online (Oh et al., 2016). Yet while there are machine learning programs being developed to collect information on users in order to recognize them, create profiles of them, and predict their behaviors, there are machine learning programs like the one developed by Bilogrevic (2016) that are simultaneously being designed to protect and preserve users’ privacy.

Another such example is in the study by Mohassel and Zhang (2017). Mohassel and Zhang (2017) use a two-server model with machine learning algorithms for the purposes of training linear regression, logistic regression and neural networks models: “in this model, in a setup phase, the data owners (clients) process, encrypt and/or secret-share their data among two non-colluding servers. In the computation phase, the two servers can train various models on the clients’ joint data without learning any information beyond the trained model” (p. 19). Their protocol as a result is 1100- 1300× faster that previous protocols developed by engineers to protect end users’ privacy in an Internet environment where scanning algorithms and constantly looking for data on users to develop their own profiles. Their protocol acts as a wall between the user and the other machine learning AI seeking...…a world of coding that in a state of constant revolution.

The recommendations based on these findings are for legislation to be developed and passed that will address the issue of using machine learning to violate privacy rights on social networks. Just as legislation emerged following the embarrassment of a U.S. Supreme Court Justice nominee in the 1980s, today’s legislators need to understand the parameters and risks of machine learning with respect to data privacy on social networks. AI can be used to protect and to violate privacy rights, and laws should be in place that dictate what is acceptable behavior on the part of machine learning developers and what is not. Until that time comes, the contest between advancing machine learning protocols to protect or to collect will only escalate.

Social networks need to be protected as personal data is left behind or made available by users, who often do not even know how to navigate the web in terms of sharing information or making data visible to machine learning algorithms. End users are some of the least informed or least prepared individuals and are typically the number one threat/risk to a system’s security. They are targeted through phishing methods and unless trained to know better can make an organization extremely vulnerable. The same goes for protecting their own data and maintaining their own privacy rights. Many users are simply uneducated about the tools available and how to use them. Thus, the other recommendation is to educate the public more broadly on how to preserve personal information on the Web.

Conclusions

Social networks are vulnerable to data harvesting algorithms and machine learning has only advanced the degree to which personal data may be gathered and interpreted for predictive purposes. The anticipated behaviors of users can be predicated upon past inferences by machine learning algorithms, but machine learning can also be used to protect the privacy of users and share only the information that the user is comfortable sharing. Machine learning protocols have even been developed that conceal data from Big Data harvesters. And yet there is no end in sight for this war over data, the conflict between what should be public and what should be private. Until legislation is passed putting an end to the…