Context-aware System for Dynamic Privacy Risk Inference

Abstract

With the rapid expansion of smart cyber-physical systems and environments, users become more and more concerned about their privacy, and ask for more involvement in the protection of their data. However, users may not be necessarily aware of the direct and indirect privacy risks they take to properly protect their privacy. In this paper, we propose a context-aware semantic reasoning system, denoted as the Privacy Oracle, capable of providing users with a dynamic overview of the privacy risks taken as their context evolves. To do so, the system continuously models, according to a proposed Semantic User Environment Modeling (SUEM) ontology, the knowledge (received by the system) about the user of interest and his surrounding cyber-physical environment. In parallel, it performs continuous reasoning over modeled information, by relying on set of privacy rules, in order to dynamically infer the privacy risks taken by the user. To validate our approach, we developed a prototype based on the semantic web tools such as OWL API, SWRL API and the inference engine Pellet. We evaluated the system performance by considering multiple use cases. Our experimental results show that the Privacy Oracle can assist users by dynamically detecting their incurred privacy risks, and by tracking, in real-time, the evolution of those risks as user context changes.

[Retracted] Design of Moral Education Management System for Higher Vocational Students Based on Multisource Sensing Data Fusion

“…Kwilinski et al [17] divided the context into three major aspects: user context, environmental context, and application context, among which user context includes an activity. Chaaya et al [18] divided the context into user context, environment context, and application context, where user context includes activity, location, and description; environment context includes time, brightness, temperature, weather, resources, and other contexts; and application context includes function, maintenance, energy, and other contexts. Mattila et al [19] divided the contextual information into three categories: natural environment, device environment, and user environment.…”

δ- Risk : Toward Context-aware Multi-objective Privacy Management in Connected Environments

“…It includes the following components: (i) context acquisition, in charge of capturing attributes from the user and her Connected/Web environments; (ii) user preferences, responsible for managing the preferences of the user; and (iii) context modeling, liable for modeling acquired attributes and the relationships that exit among them, which helps in better understanding the user context. We explored the context modeling component in previous work [2], where we proposed a generic and modular ontology for Semantic User Environment Modeling, entitled SUEM. This is motivated by the fact that adopting a semantic data model that maintains a flexible data structure becomes a fundamental requirement, especially as: (1) collected information can be heterogeneous (i.e., they have different data types and formats); (2) information can be captured from different types of data sources that could derive from both Connected environments (e.g., IoT sensor networks), and Web environments such as social networks, or any other public data source (e.g., public voting records, medical records); (3) gathered information may have different levels of granularity; and (4) performing in a dynamic environment that cannot be controlled in advance makes the system unable to control or predict the knowledge to receive, nonetheless, it must be always capable of modeling it.…”

“…We assume in this study that the privacy rules, defined by experts from the privacy community, are pre-validated (this validation is out of scope of this study). Second, the privacy risk reasoner component, which provides a semantic rule-based reasoning engine proposed in Reference [2]. This engine reasons on modeled information to dynamically infer the risks involved in the user context.…”

“…Figure 9). We execute the privacy risk inference prototype 1 proposed in our previous work [2] to infer the risks involved in this situation. Figure 10 describes the overview of risks provided to Alice, and Figure 11 illustrates the impact of energyconsump and Location attributes on these risks.…”

P-SGD: A Stochastic Gradient Descent Solution for Privacy-Preserving During Protection Transitions

“…Delivering such services requires collecting and processing massive amounts of data (e.g., location data, health data) to discover underlying patterns and trends. However, privacy concerns hinder the wider use of these data especially as data processing may give rise to serious privacy risks for individuals, such as disclosing their health conditions, habits and daily activities [4]. Consequently, balancing the trade-off between data utility and privacy protection has been subject to intense study in recent years [2,5,6,15,16].…”