Stringent regulations like General Data Protection Regulation (GDPR) mandate that an application's code-level data handling must align with its natural-language privacy policy, creating a critical auditing challenge. However, existing methods, predominantly reliant on static analysis, suffer from a critical limitation: in their pursuit of soundness via over-approximation, they exhibit "semantic blindness"—detecting what data flows exist but not why. This leads to an overwhelming volume of false positives, rendering automated auditing impractical. To bridge this gap, we introduce PriAgent, a novel framework that approaches compliance auditing as a multi-stage, AI-driven reasoning task. Instead of a monolithic model, PriAgent deploys a team of specialized agents that execute a divide-and-conquer strategy. They systematically prune the analysis space by abstracting data flows, pinpoint semantic loci critical for inspection, and perform on-demand summarization of large code blocks to ensure scalability. PriAgent leverages Retrieval-Augmented Generation (RAG) with a curated knowledge base of Android APIs, equipping agents to discern potentially non-compliant behavior from benign functionality. By correlating code-level evidence with the app's stated privacy policy, PriAgent delivers a holistic and explainable verdict for each potential violation. Our evaluations demonstrate that PriAgent significantly reduces false positives, enabling a more scalable and precise compliance audit.

Color is a powerful tool in data visualization, but for individuals with color vision deficiencies (CVD), hue can become a barrier rather than an aid. In this paper, we examine how real-world visualizations are perceived across vision profiles through three complementary studies. Study 1 assessed how normal vision participants rated 46 visualizations shown in original and simulated red/green colorblind versions. Study 2 collected matched responses from participants with diagnosed CVD. Study 3 involved in-depth interviews exploring how users interpret, adapt to, and evaluate inaccessible designs. Across studies, we find that simulations capture directional perceptual shifts but fail to reflect the interpretive breakdowns and emotional work described by real CVD users. Factor analysis reveals two dominant perceptual dimensions: functional utility and affective experience. While normal vision participants prioritize functional clarity, CVD users rely more on structural cues and emotional resonance, particularly when color is unreliable. Qualitative insights show that perceptual breakdowns occur not only in high-interference charts but also when redundant encoding or layout scaffolding is missing. We synthesize these findings and offer empirically grounded design recommendations to guide inclusive visualization practices. Our results argue that accessibility must go beyond color correction, embracing structural clarity, redundancy, and real-user validation to ensure inclusive visual communication.

Classical classification methods often fail in dynamic environments where data distributions shift over time, known as concept drift. Applications like flight delay prediction and weather forecasting require handling such dynamic data streams. Concept drift can be either virtual, affecting unconditional probability distributions, or real, affecting conditional distributions. While most research focuses on real drift, virtual drift and noise also degrade classifier performance. In this paper, we propose gaussian mixture models based on contrastive learning (GMMCL), a novel approach that integrates noise handling, contrastive learning, drift detection, and gaussian mixture models. Our method significantly enhances adaptability to noisy and drifting data streams, outperforming mainstream approaches across twelve synthetic and real-world datasets. This provides a robust solution for managing concept drift and noise in dynamic classification tasks.

Next point-of-interest (POI) recommendations have garnered significant attention in industry and academia due to their crucial role in location-based social networks (LBSNs). Recent approaches have integrated sequence and geographical data to improve recommendation accuracy. However, traditional methods do not explicitly learn user similarity, which may result in suboptimal POI prediction outcomes. To address these limitations, we propose the contrastive learning with augmented spatio-temporal(COAST) model, which more effectively utilizes user check-in sequences and geographical influences. Our approach includes five techniques for augmenting check-in records and a novel Two-Head Self-Attention Encoder (THSE) to capture spatio-temporal and structural patterns. Extensive experiments on three real-world datasets demonstrate the superiority of our model compared to existing methods.

Cross-domain sequential recommendation (CDSR) has emerged as an effective solution to address the challenges of data sparsity and cold-start issues in recommendation systems by leveraging shared knowledge across multiple domains. Nevertheless, existing methods face notable long-term challenges, such as ineffective knowledge transfer caused by distributional shifts and sparse overlapping users, insufficient modeling of temporal dynamics and intricate sequential patterns in user behavior, and suboptimal generalization across heterogeneous domains.To tackle these issues, we propose DiTAGInt, a novel diffusion-based generative network with augmented-graph embedding integration. DiTAGInt introduces a dynamic embedding fusion mechanism to harmonize domain-specific and shared-user representations, thereby enhancing generalization and alleviating rigid transfer constraints. Furthermore, it employs a diffusion-based generative module to effectively model temporal dynamics and capture complex sequential patterns in user behavior, facilitating precise user preference learning and significantly advancing recommendation accuracy. Extensive experiments conducted on three public datasets demonstrate the superiority of our method.

Stealth services have become increasingly popular due to growing demand for privacy protection. To avoid detection by active probing, many have adopted probe-resistant strategies. In this work, we design a suite of carefully crafted probes to expose their hidden vulnerabilities. Through detailed analysis of the corresponding responses, we find that despite the defensive measures implemented, certain implicit information, such as protocol stack fingerprints, can still serve as strong indicators. We present SSChecker, a detection system that combines cross-layer probing with a classification model inspired by Information Bottleneck Theory to address the data sparsity issue inherent in active probing. Our experiments on real-world datasets show that SSChecker outperforms existing methods, including leading industrial detection engines. Our findings demonstrate that current probe-resistant strategies of stealth services remain insufficient. To strengthen privacy protection, we further propose mitigation strategies that help stealth services enhance their ability.

As the scale of data on the Internet continues to increase, the management and monitoring of time series data are facing significant challenges. Efficient and stable time-series data anomaly detection methods are necessary for fields such as traffic detection, power grid operation and maintenance, financial stock market, and industry. However, there are fewer abnormal data labels in time series data, and the labeling cost is high. Traditional expert knowledge-based supervised methods have been difficult to adapt to large-scale data metric management and timely abnormal alarms. At the same time, the way based on the new neural network has an extensive time overhead when faced with massive data, and it isn’t easy to apply it in a real-time industrial environment. Therefore, we propose the TSFN model in this paper, an unsupervised method of a transformer-based self-feedback network. Which can capture timing dependencies, learn normal data distribution and improve the self-feedback ability for sensitive areas, and can be used to detect anomalies in multidimensional time series more quickly. Our experimental research on five public datasets shows that our method has fast training speed, good stability, excellent anomaly detection ability, and good generalization ability compared with the baseline method.

As the essence of urban spatiotemporal interaction systems, hubs and centers empower cities to enhance socioeconomic prosperity and sustainability. However, a city manifests a time-evolving spatial interaction network with latent temporal interactions and irregular spatial partitions. This phenomenon is termed the spatiotemporal inconsistency problem. The aggregate, single-layer network model is defective for capturing the importance of locations in such time-evolving spatial interaction systems. This article therefore proposes a novel multilayer network model based on the nature of inherent spatial and temporal dependencies of urban interactions. First, the spatial agglomeration and the temporal correlation are explicitly modeled in multilayer networks for alleviating the spatiotemporal inconsistency problem. Secondly, generalized centrality metrics from a single-layered static network to the multi-layered dynamic network are acquired in order to discover grouped hub locations over time. Lastly, the capability of the proposed method is evaluated by an empirical analysis of the taxi mobility networks of Beijing, China, from 2012 to 2017. The empirical analysis indicates that the proposed method enables the identification of typical hub locations clustered in space and stable over time. This ability is essential to understand the centrality of locations informed by noisy and inconsistent data in their spatial and temporal dimensions.