Interpretable Machine Learning for Privacy-Preserving Pervasive Systems

Abstract

Our everyday interactions with pervasive systems generate traces that capture various aspects of human behavior and enable machine learning algorithms to extract latent information about users. In this paper, we propose a machine learning interpretability framework that enables users to understand how these generated traces violate their privacy.With the emergence of connected devices (e.g., smartphones and smartmeters), pervasive systems generate growing amounts of digital traces as users undergo their everyday activities. These traces are crucial to service providers to understand their customers, to increase the degree of personalization, and enhance the quality of their services. For instance, personal digital traces stemming from public transit smartcards help transportation providers understand the commuting patterns of users; the usage statistics of home appliances can be used to improve energy efficiency; on-street cameras provide police officers with new ways of investigating crimes; content generated through mobile and wearables (e.g., posts in online social media or GPS running routes in specialized websites such as those for fitness) can be used to provide tailored content to individuals; bank transaction logs can be used to spot unusual activity in accounts.However, sharing these digital traces generated by pervasive systems with service providers might raise concerns with regards to user privacy, as the processing and analysis of these traces can surface latent information about user behaviors. Using machine learning techniques, third parties such as advertisers can identify a single individual from inadequately aggregated datasets shared by service providers either publicly or privately. The common use of ad libraries integrated directly in applications and websites further allows advertisers to collect the same raw traces as the service providers, and infer personal information about users, which can infringe on the users' privacy. In the case of location tracking libraries, these traces might reveal information

The Ethics of AI Ethics: An Evaluation of Guidelines

“…Great progress has been made in the areas of privacy, fairness or explainability. For example, many privacyfriendly techniques for the use of data sets and learning algorithms have been developed, using methods where AI systems' "sight" is "darkened" via cryptography, differential or stochastic privacy (Ekstrand, Joshaghani, and Mehrpouyan 2018;Baron and Musolesi 2017;Duchi, Jordan, and Wainwright 2013;Singla et al 2014). Nevertheless, this contradicts the observation that AI has been making such massive progress for several years precisely because of the large amounts of (personal) data available.…”

Protecting Mobile Food Diaries from Getting too Personal

“…As researchers, we usually strive to enhance utility of applications and algorithms, and often use personalisation as a tool to increase utility. While this is important, an increasing body of work has also emphasized the importance of privacy preservation and the use of less sensitive data [16,20,33,55,58,59]. Personalization and privacy preservation are at the two opposite ends of the spectrum because personalisation has typically required more personal data to provide high utility, while privacy preservation aims at providing reasonable utility from the application, while preserving privacy of users from known risks.…”

“…We believe that designing ubicomp technology for joint privacy and utility, and not only for personalisation, is important for the advancement of the field in a progressive and ethical manner. Recent literature further discusses why new privacy preservation techniques are needed by pointing out that simple anonymization techniques are no longer enough to preserve user privacy [16].…”

“…While some users might be willing to provide such data, other users would prefer to use the application without providing sensitive information, thus setting trade-offs that involve personalization, privacy, and utility, when using applications and services [28,86,90]. How this conundrum plays a role in ubiquitous computing is described in [16], which emphasizes the need for privacy-preserving systems. Moreover, according to the terms of use of several mobile health apps [4,6,11], this is exactly why they use personalization for users who provide such sensitive information, and non-personalized algorithms for users who refuse to provide such data, but still opt to use the app.…”

“…advertisers, insurance companies, etc. ), and there is no full clarity as to how companies use data [16,22,45,56,64,68,71]. According to recent reports from consumer protection agencies [1,7], this trend is not diminishing.…”