The advent of the era of big data promotes the further development of medicine, and data release is an important step in it. The existing medical data release methods mostly use the k-anonymity model as the basis for data protection. With the advancement of technology, anonymous models are progressively less resistant to consistency attacks and background knowledge attacks. In order to better protect the private information of patients, this paper makes two major contributions: (1) The method of calculating the correlation between attributes is used to ensure the validity of the data after the data is released; (2) On the basis of the previous step, combined with the difference privacy-preserving model and tree model, this paper proposes an attribute association-based differential privacy classification tree data publishing method (ACDP-Tree). In this paper, simulation experiments are carried out on real medical data sets. The experimental results show that the algorithm ensures the validity and availability of the data to a certain extent while ensuring that the patient's privacy is not leaked.
In recent years, federated GBDTs have gradually replaced traditional GBDTs, and become the focus of academic research. They are used to solve the task of structured data mining. Aiming at the problems of information leakage, insufficient model accuracy and high communication cost in the existing schemes of horizontal federated GBDTs, this paper proposes an algorithm of gradient boosting decision trees based on horizontal federated learning, that is, secure and efficient FL for GBDTs (SeFB). The algorithm uses locality sensitive hashing (LSH) to build a tree by collecting similar information of instances without exposing the original data of participants. In the stage of updating the tree, the algorithm aggregates the local gradients of all data participants and calculates the global leaf weights, so as to improve the accuracy of the model and reduce the communication cost. Finally, the experimental analysis shows that the algorithm can protect the privacy of the original data, and the communication cost is low. At the same time, the performance of the unbalanced binary data set is evaluated. The results show that SeFB algorithm compared with the existing schemes of horizontal federated GBDTs, the accuracy is improved by 2.53% on average.
Collaborative filtering recommendation systems are facing the data sparsity problem associated with interaction data, and social recommendations introduce user social information to alleviate this problem. Existing social recommendation methods cannot express the user interaction interest and social influence deeply, which limits the recommendation performance of the system. To address this problem, in this paper we propose a graph neural network social recommendation algorithm integrating multi-head attention mechanism. First, based on the user-item interaction graph and social network graph, the graph neural network is used to learn the high-order relationship between users and items and deeply extract the latent features of users and items. In the process of learning user embedding vector representation based on the social network graph, the multi-head attention mechanism is introduced to increase the importance of friends with high influence. Then, we make rating predictions for the target users according to the learned user embedding vector representation and item embedding vector. The experimental results on the Epinions dataset show that the proposed method outperforms the existing methods in terms of both Recall and Normalized Discounted Cumulative Gain.
21 September 2022 in International Journal of Distributed Sensor Networks
Location privacy refers to the individual private and sensitive location information involved in the user’s access to location services. Achieving location privacy protection has become a hot topic of research. However, existing location privacy protection schemes are susceptible to background knowledge attack, edge information attack, and homogeneity attack, on the one hand, and strict constraint on the number of neighbors, on the other hand. To address these deficiencies, a dummy location interference privacy protection algorithm for Location-Based Service location is proposed. To begin with, the dummy location candidate set is constructed based on using WordNet structure to guarantee semantic differentiation, randomly selecting offset location, and conforming to probability similarity; next, the dummy location set is filtered out by discretizing dummy locations based on the Heron formula; finally, the secure anonymity set is constructed according to the anonymity level. Experiments show that the algorithm enhances the privacy protection strength and improves the security of location privacy. Meanwhile, the communication volume and time overhead are reduced and the practicality is boosted by taking into account the sparse and dense environment of location points.
The solution of a high-resolution (HR) image corresponding to a low-resolution (LR) image is not unique in most cases. However, single-LR–single-HR supervision is widely adopted in single-image super-resolution (SISR) tasks, which leads to inflexible inference logic of the model and poor generalization ability. To improve the flexibility of model inference, we constructed a novel form of supervision, except for the ground truth (GT). Specifically, considering the structural properties of natural images, we propose using extra supervision to focus on the textural similarity of the images. As textural similarity does not account for the position information of images, a Gram matrix was constructed to break the limitations of spatial position and focus on the textural information. Besides the use of traditional perceptual loss, we propose a discriminator perceptual loss based on the two-network architecture of generative adversarial networks (GAN). The difference between the discriminator features used in this loss and the traditional visual geometry group (VGG) features is that the discriminator features can describe the relevant information from the perspective of super-resolution. Quantitative and qualitative experiments were performed to demonstrate the effectiveness of the proposed method.
As big data and data mining technology advance, research on the collection and analysis of medical data on the internet of medical things (IoMT) has gained increasing attention. Medical institutions often collect users’ signs and symptoms from their devices for analysis. However, the process of data collection may pose a risk of privacy leakage without a trusted third party. To address this issue, we propose a medical data collection based on local differential privacy and Count Sketch (MDLDP). The algorithm first uses a random sampling technique to select only one symptom for perturbation by a single user. The perturbed data is then uploaded using Count Sketch. The third-party aggregates the user-submitted data to estimate the frequencies of the symptoms and the mean extent of their occurrence. This paper theoretically demonstrates that the designed algorithm satisfies local differential privacy and unbiased estimation. We also evaluated the algorithm experimentally with existing algorithms on a real medical dataset. The results show that the MDLDP algorithm has good utility for key-value type medical data collection statistics in the IoMT.