查看更多>>摘要：Data privacy is a critical concern in the digital age. This problem has compounded with the evolution and increased adoption of machine learning (ML), which has necessitated balancing the security of sensitive information with model utility. Traditional data privacy techniques, such as differential privacy and anonymization, focus on protecting data at rest and in transit but often fail to maintain high utility for machine learning models due to their impact on data accuracy. In this article, we explore the use of synthetic data as a privacy-preserving method that can effectively balance data privacy and utility. Synthetic data is generated to replicate the statistical properties of the original dataset while obscuring identifying details, offering enhanced privacy guarantees. We evaluate the performance of synthetic data against differentially private and anonymized data in terms of prediction accuracy across various settings—different learning rates, network architectures, and datasets from various domains. Our findings demonstrate that synthetic data maintains higher utility (prediction accuracy) than differentially private and anonymized data. The study underscores the potential of synthetic data as a robust privacy-enhancing technology (PET) capable of preserving both privacy and data utility in machine learning environments.
查看更多>>摘要：Graph anomaly detection is a challenging task in graph data mining, aiming to recognize unconventional patterns within a network. Recently, there has been increasing attention on graph anomaly detection based on contrastive learning due to its high adaptability to the sample imbalance problem. However, most existing work typically focuses on the contrast of local views while neglecting global comparison information, leading to suboptimal performance. To address this issue, we introduce a new multiscale contrastive self-supervised learning framework for graph anomaly detection (GADMCLG). Our approach incorporates local-level contrasts involving node–node and node–subgraph contrast, and global-level subgraph–subgraph contrast. The former mines localized abnormal information, while the latter is intended to capture global anomalous patterns. Specifically, our proposed subgraph–subgraph contrast adopts the h-order neighbor subgraph sampling instead of augmented subgraphs through edge perturbation. This sampling strategy ensures a comprehensive observation of the neighborhood surrounding the target node, thereby mitigating the introduction of extraneous noise and providing interpretability for the detected results. Furthermore, we incorporate a subgraph centralization technique to reduce the bias caused by the absolute position of subgraphs in the attribute space, which enhances the model's ability to identify anomalies at different scales. Extensive experimental results on six real-world datasets demonstrate the effectiveness of our method and its superiority compared with state-of-the-art approaches.
查看更多>>摘要：Ethereum, as one of the most active cryptocurrency trading platforms, has garnered significant academic interest due to its transparent and accessible transaction data. In recent years, phishing scams have emerged as a serious criminal activity on Ethereum. Although most studies model Ethereum account transactions as networks and analyze them using traditional machine learning or network representation learning techniques, these approaches often rely solely on the latest static transaction records or use manually designed features while neglecting transaction histories, thus failing to fully capture the dynamic interactions and potential trading patterns between accounts. This article introduces an innovative multiperspective cascaded dynamic graph neural network model named DMPCG, which extracts phishing transaction data from authoritative databases like blockchain explorers to construct transaction network graphs. The model elevates the analysis from the microscopic features of nodes to the macroscopic dynamics of the entire network, integrating the attributes of static snapshot graphs with the evolution of dynamic trading networks, significantly enhancing the accuracy of phishing detection. Experimental results demonstrate that the DMPCG method achieves an impressive precision of 92.6% and an F1-score of 90.9%, outperforming existing baseline models and traditional subgraph sampling techniques.
查看更多>>摘要：The computational analysis of big data has revolutionized social science research, offering unprecedented insights into societal behaviors and trends through digital data from online sources. However, existing tools often face limitations such as technical complexity, single-source dependency, and a narrow range of analytical capabilities, hindering accessibility and effectiveness. This article introduces DataPoll, an end-to-end big data analysis platform designed to democratize computational social science research. DataPoll simplifies data collection, analysis, and visualization, making advanced analytics accessible to researchers of diverse expertise. It supports multisource data integration, innovative analytical features, and interactive dashboards for exploratory and comparative analyses. By fostering collaboration and enabling the integration of new data sources and analysis methods, DataPoll represents a significant advancement in the field. A comprehensive case study on the Ukrainian–Russian conflict demonstrates its capabilities, showcasing how DataPoll can yield actionable insights into complex social phenomena. This tool empowers researchers to harness the potential of big data for impactful and inclusive research.
查看更多>>摘要：Accurately predicting the future trajectories of traffic agents is paramount for autonomous unmanned systems, such as self-driving cars and mobile robotics. Extracting abundant temporal and social features from trajectory data and integrating the resulting features effectively pose great challenges for predictive models. To address these issues, this article proposes a novel multibranch attentive transformer (MBAT) trajectory prediction network for traffic agents. Specifically, to explore and reveal diverse correlations of agents, we propose a decoupled temporal and spatial feature learning module with multibranch to extract temporal, spatial, as well as spatiotemporal features. Such design ensures each branch can be specifically tailored for different types of correlations, thus enhancing the flexibility and representation ability of features. Besides, we put forward an attentive transformer architecture that simultaneously models the complex correlations possibly occurring in historical and future timesteps. Moreover, the temporal, spatial, and spatiotemporal features can be effectively integrated based on different types of attention mechanisms. Empirical results demonstrate that our model achieves outstanding performance on public ETH, UCY, SDD, and INTERACTION datasets. Detailed ablation studies are conducted to verify the effectiveness of the model components.
NETL
NSTL
IEEE