Local differential privacy (LDP) provides strict privacy guarantee in a distributed environment. Recent studies demonstrated that LDP protocols are vulnerable to data poisoning attacks where an attacker can manipulate the perturbed result on the local side and send bogus data to skew the final estimate on the server. Unfortunately, existing attack detections do not create an effective attack indicator and rely on particular characteristics of LDP protocols. As a result, they typically exhibit limited detection performance. In this paper, we use log-likelihood as the attack indicator and propose a chain-style detection to enhance the detection effectiveness, in which the attack impact could propagate along the chain and exhibit clear anomaly signal even under stealthy attack scenarios. The experimental results show that our detection consistently outperforms the existing methods. Using four datasets containing categorical and numerical data separately, our detection achieves an F1 score exceeding 96% in most cases. It even remains above 0.9 under stealthy attack settings, outperforming the state-of-the-art detection by up to 0.25.

With the rapid advancement and widespread ap plication of the graph neural network (GNN), the collaborative graph learning, in which multiple parties collaboratively construct a GNN model using their respective graph data, has attracted increasing attention. However, this paradigm also raises significant privacy concerns, as both nodes and edges may contain sensitive personal information, while existing privacy preserving schemes often come at the cost of degraded model performance or substantial system overhead. Therefore, this paper proposes an efficient and privacy-preserving collaborative, and Hui Li, Member, IEEE, Xiaoyu Kou Social Platform learning framework on vertically partitioned graph data, dubbed Plog. Specifically, we first design a decomposition algorithm to split the sparse adjacency matrix into the summation of multiple independent permutations, which are lightweight, parallelizable, and well-suited for secure multi-party computation. Building on this, a weighted oblivious batch permutation protocol is carefully customized based on correlated randomness to securely and efficiently compute adjacency matrix multiplications, addressing the core efficiency bottleneck in GNN inference and training. The selective security of Plog is formally verified under the ideal-real paradigm. Extensive experimental results on three real world datasets demonstrate that compared to the state-of-the art scheme, Plog can reduce online communication rounds by 46% and achieve a 1.73× speedup in the overall inference and training time.

In the era of cloud computing and AI, data owners outsource ubiquitous vectors to the cloud, which furnish approx-imate k-nearest neighbors (k-ANNS) services to users. To protect data privacy against the untrusted server, privacy-preserving k-ANNS (PP-ANNS) on vectors has been a fundamental and urgent problem. However, existing PP-ANNS solutions fall short of meeting the requirements of data privacy, efficiency, accuracy, and minimal user involvement concurrently. To tackle this challenge, we introduce a novel solution that primarily executes PP-ANNS on a single cloud server to avoid the heavy communication overhead between the cloud and the user. To ensure data privacy, we introduce a novel encryption method named distance comparison encryption, facilitating secure, efficient, and exact distance comparisons. To optimize the trade-off between data privacy and search performance, we design a privacy-preserving index that combines the state-of-the-art k-ANNS method with an approximate distance computation method. Then, we devise a search method using a filter-and - refine strategy based on the index. Moreover, we provide the security analysis of our solution and conduct extensive experiments to demonstrate its superiority over existing solutions. Based on our experimental results, our method accelerates PP-ANNS by up to 3 orders of magnitude compared to state-of-the-art methods, while not compromising the accuracy.

Federated Learning (FL) is a promising privacy-preserving machine learning paradigm that allows data owners to collaboratively train models while keeping their data localized. Despite its potential, FL faces challenges related to the trustworthiness of both clients and servers, particularly against curious or malicious adversaries. In this paper, we introduce a novel framework namedFederatedLearning with Low-DimensionalUpdateRepresentation andProximity-Based defense (FLURP), designed to address privacy preservation and resistance to Byzantine attacks in distributed learning environments. FLURP employs$\mathsf {LinfSample}$method, enabling clients to compute the$l_{\infty }$norm across sliding windows of updates, resulting in a Low-Dimensional Update Representation (LUR). Calculating the shared distance matrix among LURs, rather than updates, significantly reduces the overhead of Secure Multi-Party Computation (SMPC) by three orders of magnitude while effectively distinguishing between benign and poisoned updates. Additionally, FLURP integrates a privacy-preserving proximity-based defense mechanism utilizing optimized SMPC protocols to minimize communication rounds. Our experiments demonstrate FLURP's effectiveness in countering Byzantine adversaries with low communication and runtime overhead. FLURP offers a scalable framework for secure and reliable FL in distributed environments, facilitating its application in scenarios requiring robust data management and security.

With the increasing severity of user privacy leaks in online social networks (OSNs), existing privacy protection technologies have difficulty meeting the diverse privacy protection needs of users. Therefore, privacy‐aware (PA) for the text data that users post on OSNs has become a current research focus. However, most existing PA algorithms for OSN users only provide the types of privacy disclosures rather than the specific locations of disclosures. Furthermore, although named entity recognition (NER) technology can extract specific locations of privacy text, it has poor recognition performance for nested and interest privacy. To address these issues, this paper proposes a PA framework based on the extraction of OSN privacy information content. The framework can automatically perceive the privacy information shared by users in OSNs and accurately locate which parts of the text are leaking sensitive information. Firstly, we combine the roformerBERT model, BI_LSTM model, and global_pointer algorithm to construct a direct privacy entity recognition (DPER) model for solving the specific privacy location recognition and entity nesting problems. Secondly, we use the roformerBERT model and UniLM framework to construct an interest privacy inference (IPI) model for interest recognition and to generate interpretable text that supports this interest. Finally, we constructed a dataset of 13,000 privacy‐containing texts for experimentation. Experimental results show that the overall accuracy of the DPER model can reach 91.80%, while that of the IPI model can reach 98.3%. Simultaneously, we compare the proposed model with recent methods. The analysis of the results indicates that the proposed model exhibits better performance than previous methods.

Online information diffusion generates huge volumes of social activities (eg. tweets, retweets posts, comments, likes) among individuals. Existing information diffusion modeling techniques are oblivious to conformity of individuals during the diffusion process, a fundamental human trait according to social psychology theories. Intuitively, conformity captures the extent to which an individual complies with social norms or expectations. In this paper, we present a novel framework called chassis to characterize online information diffusion by bridging classical information diffusion model with conformity from social psychology. To this end, we first extend "Hawkes Process", a well-known statistical technique utilized to model information diffusion, to quantitatively capture two flavors of conformity, informational conformity and normative conformity, hidden in activity sequences. Next, we present a novel semi-parametric inference approach to learn the proposed model. Experimental study with real-world datasets demonstrates the superiority of chassis to state-of-the-art conformity-unaware information diffusion models.