Data sharing facilitates the integration and in-depth exploration of cross-domain data, thereby fostering innovative research and model development. However, privacy leakage emerges as a critical barrier to the sharing and circulating of such data. Existing privacy-preserving technologies face challenges in handling complex scenarios involving multiple participants due to the following reasons: 1) Divergent privacy permission. Data sharing is constrained by various privacy limitations, necessitating the consideration of privacy permissions across different domains, akin to a cross-border process. 2) High collaboration cost. Collaboration among multiple domains to determine the privacy constraint and sharing ways incur additional costs. 3) Large noise magnitude. Traditional privacy techniques to protect the privacy of a single domain using local differential privacy (LDP) may introduce excessive noise, thereby reducing data utility. Drawing inspiration from the cross-border visa issuance process, we present an innovative framework called PriVisa for enabling privacy-preserving data sharing across different domains. It consists of four key modules to overcome the mentioned challenges: the hybrid pattern, optimized sharing path construction, personalized grouping, and LDP-based perturbation. 1) The hybrid pattern for coordination among organizations, considering authentication, privacy constraints, and sharing methods. 2) The optimized sharing path construction using a privacy constraint hierarchy tree to maximize data utility while adhering to privacy requirements. 3) The feature similarity grouping and perturbing mechanism satisfying LDP to protect privacy and optimize data utility. The theoretical and experimental validation confirms PriVisa’s effectiveness in addressing divergent privacy constraints and promoting data utility in cross-domain data sharing.

Nowadays, the emergence of online trading greatly facilitates people’s life. Meanwhile, online trading also brings hidden dangers, such as online fraudulent trading. To solve the issue, researchers have proposed many different detection models. However, in actual business scenarios, fraudulent transactions usually only account for a small portion of normal transactions, resulting in extremely imbalanced data. Besides, the concealment of fraud is reflected in that the fraudsters are imitating the normal transactions of users, posing a huge challenge for fraudulent transaction detection modeling. Inspired by generative adversarial networks (GANs), we propose a GAN-based framework to detect online banking fraud on extremely imbalanced data, called BalanceGAN. A fraud detection model is first pretrained using the data generated by the generator and then the model is fine-tuned using transfer learning on real-world datasets, by using this approach to address data imbalances. Compared with the conventional methods for solving imbalanced data, our BalanceGAN can avoid over-fitting of the model relatively, experiments on two real datasets show that our BalanceGAN has more than 10% performance improvement in Precision and Recall.

The possibility for public attributes to disclose private information has caused widespread concern. Traditional privacy-preserving approaches have two limitations: 1) Approaches based on data anonymization or distortion often lead to poor utility-privacy trade-offs, and 2) approaches based on data encryption face heavy computational costs. These problems have prompted calls for an effective privacy-preserving framework that provides adequate privacy guarantees while maintaining good data utility. Inspired by denoising autoencoders, in this paper, we regard the information about privacy attributes contained in the public attributes as a kind of noise and design an ex ante privacy-preserving model called the Mutual Information Autoencoder (MIAE), which reconstructs the loss function of the original autoencoder by combining reconstruction errors and mutual information, and we introduce a trade-off coefficient to achieve utility-privacy trade-offs. To elucidate the superiority of the proposed model, we consider utility-privacy trade-offs with the expected distortion function as a metric of data utility and the joint mutual information as a metric of privacy disclosure, and then, we construct a convex optimization problem with multiple constraints based on rate-distortion theory. From an information theory perspective, we provide a lower bound for privacy disclosure with utility guarantees. Elaborate experiments over a real-world dataset reveal that as the level of expected distortion increases, the achievable bound obtained by MIAE exhibits a trend similar to that of the information-theoretic bound. When the expected distortion surpasses 2.2, the achievable bound obtained by MIAE also converges to 0, and the maximum gap between the achievable bound obtained by MIAE and the information-theoretic bound is no more than 1.4. Compared to existing models, MIAE can provide a tighter achievable bound and achieve good utility-privacy trade-offs.

Local differential privacy (LDP) is a prevalent measure of privacy protection as it provides rigorous privacy guarantees and has been widely studied for statistical analysis, especially in frequency estimation. As a representative LDP-enabled frequency estimation algorithm, Google'sRandomized Aggregation Privacy-Preserving Ordinal Response(RAPPOR) has been put into practice. However, it achieves sub-optimal utility due to the following limitations. Firstly, the adoption of the MD5 hash function inevitably results in the hash collision. Secondly, the application of the randomized response technique leads to randomness. To improve the practical effectiveness of RAPPOR and the utility of frequency-based services, we propose an LDP-enabled frequency estimation method called PK-RAPPOR, in which we devise an effective re-encoding hash function (RE-HF) incorporating prior knowledge (PK) about the rough frequency ranking of items. RE-HF divides items into several cohorts based on the PK and generates a unique hash value set for each item. Compared with the original RAPPOR, the hash collision can be eliminated for items from different cohorts, and the effect of randomness can be decreased by the overlapping of items from the same cohorts. We validate our proposed method with theoretical analysis and demonstrate its effectiveness with experiments on both synthetic and real-world datasets.

Big data can help with providing valuable perceptions into business activities and disclosing the potential benefits. Advances in machine learning and deep learning technologies make it easier to achieve significant performance in a wide range of domains from city planning and marketing analysis to credit evaluation and identity theft detection. However, it still requires great efforts in selecting efficient learning algorithms and precise model parameters that are deemed confidential in the light of experience. Also worth noting is that there is a fundamental gap between impracticable business requirements and the available value of data reflected. The data holder or data service provider may not have a clear understanding of data interference. The solution to these two problems depends on the capability of predicting the data utility in advance, which raises a fundamental question: to what degree is the data utility predictable? In this work, we present a primary analytical framework for information-theoretic bounds of data utility and utilize the current state-of-the-art and representative algorithms to obtain the achievable lower bounds on a real-world data set. The gap between theoretical upper bounds and achievable lower bounds indicates that the achievable lower bounds can still be optimized for performance.