The extensive use of smartphones and wearable devices has facilitated many useful applications. For example, with Global Positioning System (GPS)-equipped smart and wearable devices, many applications can gather, process, and share rich metadata, such as geolocation, trajectories, elevation, and time. For example, fitness applications, such as Runkeeper and Strava, utilize the information for activity tracking and have recently witnessed a boom in popularity. Those fitness tracker applications have their own web platforms and allow users to share activities on such platforms or even with other social network platforms. To preserve the privacy of users while allowing sharing, several of those platforms may allow users to disclose partial information, such as the elevation profile for an activity, which supposedly would not leak the location of the users. In this work, and as a cautionary tale, we create a proof of concept where we examine the extent to which elevation profiles can be used to predict the location of users. To tackle this problem, we devise three plausible threat settings under which the city or borough of the targets can be predicted. Those threat settings define the amount of information available to the adversary to launch the prediction attacks. Establishing that simple features of elevation profiles, e.g., spectral features, are insufficient, we devise both natural language processing (NLP)-inspired text-like representation and computer vision-inspired image-like representation of elevation profiles, and we convert the problem at hand into text and image classification problem. We use both traditional machine learning- and deep learning-based techniques and achieve a prediction success rate ranging from 59.59% to 99.80%. The findings are alarming, highlighting that sharing elevation information may have significant location privacy risks.

Enabling users to push the physical world’s limits, augmented and virtual reality platforms opened a new chapter in perception. Novel immersive experiences resulted in the emergence of new interaction methods for virtual environments, which came with unprecedented security and privacy risks. This paper presents a keylogging inference attack to infer user inputs typed with in-air tapping keyboards. We observe that hands follow specific patterns when typing in the air and exploit this observation to carry out our attack. Starting with three plausible attack scenarios where the adversary obtains the hand trace patterns of the victim, we build a pipeline to reconstruct the user input. Our attack pipeline takes the hand traces of the victim as an input and outputs a set of input inferences ordered from the best to worst. Through various experiments, we showed that our inference attack achieves a pinpoint accuracy ranging from 40% to 87% within at most the top-500 candidate reconstructions. Finally, we discuss countermeasures, while the results presented provide a cautionary tale of the security and privacy risk of the immersive mobile technology.

The privacy policy statements are the primary mean for service providers to inform Internet users about their data collection and use practices, although they often are long and lack a specific structure. In this work, we introduce TLDR, a pipeline that employs various deep representation techniques for normalizing policies through learning and modeling, and an automated ensemble classifier for privacy policy classification. TLDR advances the state-of-the-art by (i) categorizing policy contents into nine privacy policy categories with high accuracy, (ii) detecting missing information in privacy policies, and (iii) significantly reducing policy reading time and improving understandability by users.

The extensive use of smartphones and wearable devices has facilitated many useful applications. For example, with Global Positioning System (GPS)-equipped smart and wearable devices, many applications can gather, process, and share rich metadata, such as geolocation, trajectories, elevation, and time. For example, fitness applications, such as Runkeeper and Strava, utilize information for activity tracking, and have recently witnessed a boom in popularity. Those fitness tracker applications have their own web platforms, and allow users to share activities on such platforms, or even with other social network platforms. To preserve privacy of users while allowing sharing, several of those platforms may allow users to disclose partial information, such as the elevation profile for an activity, which supposedly would not leak the location of the users. In this work, and as a cautionary tale, we create a proof of concept where we examine the extent to which elevation profiles can be used to predict the location of users. To tackle this problem, we devise three plausible threat settings under which the city or borough of the targets can be predicted. Those threat settings define the amount of information available to the adversary to launch the prediction attacks. Establishing that simple features of elevation profiles, e.g., spectral features, are insufficient, we devise both natural language processing (NLP)-inspired text-like representation and computer vision-inspired image-like representation of elevation profiles, and we convert the problem at hand into text and image classification problem. We use both traditional machine learning-and deep learning-based techniques, and achieve a prediction success rate ranging from 59.59% to 95.83%. The findings are alarming, and highlight that sharing elevation information may have significant location privacy risks.

The main goal of this study is to investigate the robustness of graph-based Deep Learning (DL) models used for Internet of Things (IoT) malware classification against Adversarial Learning (AL). We designed two approaches to craft adversarial IoT software, including Off-the-Shelf Adversarial Attack (OSAA) methods, using six different AL attack approaches, and Graph Embedding and Augmentation (GEA). The GEA approach aims to preserve the functionality and practicality of the generated adversarial sample through a careful embedding of a benign sample to a malicious one. Our evaluations demonstrate that OSAAs are able to achieve a misclassification rate (MR) of 100%. Moreover, we observed that the GEA approach is able to misclassify all IoT malware samples as benign.