Training deep learning models on sensitive user data has raised increasing privacy concerns in many areas. Federated learning is a popular approach for privacy protection that collects the local gradient information instead of raw data. One way to achieve a strict privacy guarantee is to apply local differential privacy into federated learning. However, previous works do not give a practical solution due to two issues. First, the range difference of weights in different deep learning model layers has not been explicitly considered when applying local differential privacy mechanism. Second, the privacy budget explodes due to the high dimensionality of weights in deep learning models and many query iterations of federated learning. In this paper, we proposed a novel design of local differential privacy mechanism for federated learning to address the abovementioned issues. It makes the local weights update differentially private by adapting to the varying ranges at different layers of a deep neural network, which introduces a smaller variance of the estimated model weights, especially for deeper models. Moreover, the proposed mechanism bypasses the curse of dimensionality by parameter shuffling aggregation. A series of empirical evaluations on three commonly used datasets in prior differential privacy works, MNIST, Fashion-MNIST and CIFAR-10, demonstrate that our solution can not only achieve superior deep learning performance but also provide a strong privacy guarantee at the same time.

Voice input users’ speech recordings are being collected by service providers and shared with third parties, who may abuse users’ voiceprints, identify them by voice, and learn their sensitive speech content. In this work, we designSpeech Sanitizerto perturb users’ speech recordings so that the sanitized speech can be safely shared with third parties. First, we desensitize speech content by identifying sensitive words, localizing them in the audio using DTW-based keyword spotting, and substituting them with safe words. Both common and personalized sensitive words are identified and replaced. Then, we anonymize users’ voiceprints with a carefully designed voice conversion mechanism that is resistant to de-anonymization attacks. Meanwhile, we try to preserve the utility of the sanitized speech, measured by the accuracy of speech recognition performed on it. We implement Speech Sanitizer and present extensive experimental results that validate the effectiveness and efficiency of our algorithms. It is demonstrated that we are able to reduce the chance of a user's voice being identified from 50 people by 83.7 percent while keeping the drop of speech recognition accuracy within 19.1 percent. We can also easily relax the privacy level to improve speech recognition accuracy.

Privacy has become one of the major concerns in cloud video surveillance. Privacy protection of the surveillance videos strive to protect users' privacy information without hampering regular security tasks of the surveillance, meanwhile retains the system's high accuracy and efficiency. The current state of the art in protecting the video privacy is mainly realized through Privacy Region Protection, which only protects the privacy regions while keeps the non-privacy regions visually intact so that processing in the cloud is still feasible. However, the problem of determining the privacy regions has been ignored and not properly addressed. In this paper, we propose a novel notion - concept graph, and with the aid of that, we develop our system - PatronuS to determine the privacy regions subject to satisfying both privacy and security requirements. We further propose an event distilling model and a privacy inference model to assist in determining specific privacy regions. And we evaluate PatronuS in real-world settings and demonstrate its efficiency in privacy protection without degrading system's surveillance functionality.

Machine learning (ML) has shown its impressive performance in the modern world, and many corporations leverage the technique of machine learning to improve their service quality, e.g., Facebook's DeepFace. Machine learning models with a collection of private data being processed by a training algorithm are deemed to be increasingly confidential. Confidential models are typically trained in a centralized cloud server but publicly accessible. ML-as-a-service (MLaaS) system is one of running examples, where users are allowed to access trained models and are charged on a pay-per-query basis.

Sign language is a natural and fully-formed communication method for deaf or hearing-impaired people. Unfortunately, most of the state-of-the-art sign recognition technologies are limited by either high energy consumption or expensive device costs and have a difficult time providing a real-time service in a daily-life environment. Inspired by previous works on motion detection with wearable devices, we propose Sign Speaker - a real-time, robust, and user-friendly American sign language recognition (ASLR) system with affordable and portable commodity mobile devices. SignSpeaker is deployed on a smartwatch along with a smartphone; the smartwatch collects the sign signals and the smartphone outputs translation through an inbuilt loudspeaker. We implement a prototype system and run a series of experiments that demonstrate the promising performance of our system. For example, the average translation time is approximately $1.1$ seconds for a sentence with eleven words. The average detection ratio and reliability of sign recognition are 99.2% and 99.5%, respectively. The average word error rate of continuous sentence recognition is 1.04% on average.

Data quality assessment is vital for many information services ranging from sensor networks to smart city systems. The current data quality assessments, however, are often derived from intrinsic data characteristics, disconnected from specific application contexts, or are not applicable or efficient for large datasets. In this work, we propose a novel task-oriented data quality assessment framework, which balances between the intrinsic and contextual quality. We carefully craft the assessment metrics, quantify them, and fuse them to rank candidate datasets by quality given specific tasks. To improve the system efficiency, two fast calculation algorithms are designed to quantify the relationship between datasets and the task, and the distribution of data items. We conduct extensive evaluations on six public image datasets (with 460, 247 images in total) and four text document datasets (with 37, 372 documents in total) to evaluate the efficacy and efficiency of our design. Experimental results show that our algorithms can save about 90% computing time with little accuracy loss which validates the feasibility and effectiveness of our framework for large datasets.

Social network data is widely shared, transferred and published for research purposes and business interests, but it has raised much concern on users' privacy. Even though users' identity information is always removed, attackers can still de-anonymize users with the help of auxiliary information. To protect against de-anonymization attack, various privacy protection techniques for social networks have been proposed. However, most existing approaches assume specific and restrict network structure as background knowledge and ignore semantic level prior belief of attackers, which are not always realistic in practice and do not apply to arbitrary privacy scenarios. Moreover, the privacy inference attack in the presence of semantic background knowledge is barely investigated. To address these shortcomings, in this work, we introduce knowledge graphs to explicitly express arbitrary prior belief of the attacker for any individual user. The processes of de-anonymization and privacy inference are accordingly formulated based on knowledge graphs. Our experiment on data of real social networks shows that knowledge graphs can power de-anonymization and inference attacks, and thus increase the risk of privacy disclosure. This suggests the validity of knowledge graphs as a general effective model of attackers' background knowledge for social network attack and privacy preservation.

In this paper, a set of accountable protocols denoted as AccountTrade is proposed for big data trading among dishonest consumers. For achieving a secure big data trading environment, AccountTrade achieves book-keeping ability and accountability against dishonest consumers throughout the trading (i.e., buying and selling) of datasets. We investigate the consumers' responsibilities in the dataset trading, then we design AccountTrade to achieve accountability against dishonest consumers that are likely to deviate from the responsibilities. Specifically, a uniqueness index is defined and proposed, which is a new rigorous measurement of the data uniqueness for this purpose. Furthermore, several accountable trading protocols are presented to enable data brokers to blame the misbehaving entities when misbehavior is detected. The accountability of AccountTrade is formally defined, proved, and evaluated by an automatic verification tool as well as extensive simulation with real-world datasets. Our evaluation shows that AccountTrade incurs at most 10-kB storage overhead per file, and it is capable of 8-1000 concurrent data upload requests per server.

Previous works on social network de-anonymization focus on designing accurate and efficient de-anonymization methods. We attempt to investigate the intrinsic relationship between the attacker’s knowledge and the expected de-anonymization gain. A common intuition is that more knowledge results in more successful de-anonymization. However, our analysis shows this is not necessarily true if the attacker uses the full background knowledge for de-anonymization. Our findings leave intriguing implications for the attacker to make better use of the background knowledge for de-anonymization and for the data owners to better measure the privacy risk when releasing their data to third parties.

Privacy-preserving data publishing has been a heated research topic in the last decade. Numerous ingenious attacks on users' privacy and defensive measures have been proposed for the sharing of various data, varying from relational data, social network data, spatiotemporal data, to images and videos. Speech data publishing, however, is still untouched in the literature. To fill this gap, we study the privacy risk in speech data publishing and explore the possibilities of performing data sanitization to achieve privacy protection while preserving data utility simultaneously. We formulate this optimization problem in a general fashion and present thorough quantifications of privacy and utility. We analyze the sophisticated impacts of possible sanitization methods on privacy and utility, and also design a novel method - key term perturbation for speech content sanitization. A heuristic algorithm is proposed to personalize the sanitization for speakers to restrict their privacy leak (p-leak limit) while minimizing the utility loss. The simulations of linkage attacks and sanitization on real datasets validate the necessity and feasibility of this work.

Deep Learning has shown promising performance in a variety of pattern recognition tasks owning to large quantities of training data and complex structures of neural networks. However conventional deep neural network (DNN) training involves centrally collecting and storing the training data, and then centrally training the neural network, which raises much privacy concerns for the data producers. In this paper, we study how to enable deep learning without disclosing individual data to the DNN trainer. We analyze the risks in conventional deep learning training, then propose a novel idea - Crowdlearning, which decentralizes the heavy- load training procedure and deploys the training into a crowd of computation-restricted mobile devices who generate the training data. Finally, we propose SliceNet, which ensures mobile devices can afford the computation cost and simultaneously minimize the total communication cost. The combination of Crowdlearning and SliceNet ensures the sensitive data generated by mobile devices never leave the devices, and the training procedure will hardly disclose any inferable contents. We numerically simulate our prototype of SliceNet which crowdlearns an accurate DNN for image classification, and demonstrate the high performance, acceptable calculation and communication cost, satisfiable privacy protection, and preferable convergence rate, on the benchmark DNN structure and dataset.

We are speeding toward a not-too-distant future when we can perform human-computer interaction using solely our voice. Speech recognition is the key technology that powers voice input, and it is usually outsourced to the cloud for the best performance. However, user privacy is at risk because voiceprints are directly exposed to the cloud, which gives rise to security issues such as spoof attacks on speaker authentication systems. Additionally, it may cause privacy issues as well, for instance, the speech content could be abused for user profiling. To address this unexplored problem, we propose to add an intermediary between users and the cloud, named VoiceMask, to anonymize speech data before sending it to the cloud for speech recognition. It aims to mitigate the security and privacy risks by concealing voiceprints from the cloud. VoiceMask is built upon voice conversion but is much more than that; it is resistant to two de-anonymization attacks and satisfies differential privacy. It performs anonymization in resource-limited mobile devices while still maintaining the usability of the cloud-based voice input service. We implement VoiceMask on Android and present extensive experimental results. The evaluation substantiates the efficacy of VoiceMask, e.g., it is able to reduce the chance of a user's voice being identified from 50 people by a mean of 84%, while reducing voice input accuracy no more than 14.2%.

We propose AccountTrade, a set of accountable protocols, for big data trading among dishonest consumers. To secure the big data trading environment, our protocols achieve book-keeping ability and accountability against dishonest consumers who may misbehave throughout the dataset transactions. Specifically, we study the responsibilities of the consumers in the dataset trading and design AccountTrade to achieve accountability against the dishonest consumers who may try to deviate from their responsibilities. Specifically, we propose uniqueness index, a new rigorous measurement of the data uniqueness, as well as several accountable trading protocols to enable data brokers to blame the dishonest consumer when misbehavior is detected. We formally define, prove, and evaluate the accountability of our protocols by an automatic verification tool as well as extensive evaluation in real-world datasets. Our evaluation shows that AccountTrade incurs negligible constant storage overhead per file (<;10KB), and it is able to handle 8-1000 concurrent data uploading per server depending on the data types.

Tons of online user behavior data are being generated every day on the booming and ubiquitous Internet. Growing efforts have been devoted to mining the abundant behavior data to extract valuable information for research purposes or business interests. However, online users' privacy is thus under the risk of being exposed to third-parties. The last decade has witnessed a body of research works trying to perform data aggregation in a privacy-preserving way. Most of existing methods guarantee strong privacy protection yet at the cost of very limited aggregation operations, such as allowing only summation, which hardly satisfies the need of behavior analysis. In this paper, we propose a scheme PPSA, which encrypts users' sensitive data to prevent privacy disclosure from both outside analysts and the aggregation service provider, and fully supports selective aggregate functions for online user behavior analysis while guaranteeing differential privacy. We have implemented our method and evaluated its performance using a trace-driven evaluation based on a real online behavior dataset. Experiment results show that our scheme effectively supports both overall aggregate queries and various selective aggregate queries with acceptable computation and communication overheads.

We propose a graph-based framework for privacy preserving data publication, which is a systematic abstraction of existing anonymity approaches and privacy criteria. Graph is explored for dataset representation, background knowledge specification, anonymity operation design, as well as attack inferring analysis. The framework is designed to accommodate various datasets including social networks, relational tables, temporal and spatial sequences, and even possible unknown data models. The privacy and utility measurements of the anonymity datasets are also quantified in terms of graph features. Our experiments show that the framework is capable of facilitating privacy protection by different anonymity approaches for various datasets with desirable performance.

Social network data is widely shared, transferred and published for research purposes and business interests, but it has raised much concern on users' privacy. Even though users' identity information is always removed, attackers can still de-anonymize users with the help of auxiliary information. To protect against de-anonymization attack, various privacy protection techniques for social networks have been proposed. However, most existing approaches assume specific and restrict network structure as background knowledge and ignore semantic level prior belief of attackers, which are not always realistic in practice and do not apply to arbitrary privacy scenarios. Moreover, the privacy inference attack in the presence of semantic background knowledge is barely investigated. To address these shortcomings, in this work, we introduce knowledge graphs to explicitly express arbitrary prior belief of the attacker for any individual user. The processes of de-anonymization and privacy inference are accordingly formulated based on knowledge graphs. Our experiment on data of real social networks shows that knowledge graphs can strengthen de-anonymization and inference attacks, and thus increase the risk of privacy disclosure. This suggests the validity of knowledge graphs as a general effective model of attackers' background knowledge for social network privacy preservation.

The rapid information propagation facilitates our work and life without precedent in history, but it has tremendously exaggerated the risk and consequences of privacy invasion. Today's attackers are becoming more and more powerful in gathering personal information from many sources and mining these data to further uncover users' privacy. A great number of previous works have shown that, with adequate background knowledge, attackers are even able to infer sensitive information that is not revealed to anyone malicious before. In this paper, we model the attacker's knowledge using a knowledge graph and formally define the privacy inference problem. We show its #P-hardness and design an approximation algorithm to perform privacy inference in an iterative fashion, which also reflects real-life network evolution. The simulations on two data sets demonstrate the feasibility and efficacy of privacy inference using knowledge graphs.

Online user behavior analysis is becoming increasingly important, and offers valuable information to analysts for developing better e-commerce strategies. However, it also raises significant privacy concerns. Recently, growing efforts have been devoted to protecting the privacy of individuals while data aggregation is performed, which is a critical operation in behavior analysis. Unfortunately, existing methods allow very limited aggregation over user data, such as allowing only summation, which hardly satisfies the need of behavior analysis. In this paper, we propose a scheme PPSA, which encrypts users' sensitive data to prevent privacy leakage from both analysts and the aggregation service provider, and fully supports selective aggregate functions for differentially private data analysis. We have implemented our design and evaluated its performance using a trace-driven evaluation based on an online behavior dataset. Evaluation results show that our scheme effectively supports various selective aggregate queries with acceptable computation and communication overheads.