horizon scanning

This study addresses critical challenges in infectious disease prospective surveillance—namely, delayed risk signal detection, inadequate multi-source data fusion, and sluggish early-warning response—by proposing an AI-driven prospective governance framework. Methodologically, it integrates natural language processing, streaming big data analytics, and dynamic predictive modeling to construct a real-time, closed-loop system for global public-data monitoring, risk assessment, and scenario simulation, augmented with AI ethics evaluation and collaborative governance mechanisms. Key contributions include: (1) the first implementation of minute-level intelligent detection and multidimensional causal attribution of early infectious disease risk signals; (2) empirical validation demonstrating an average 5.3-day advance in outbreak预警 (p < 0.01), significantly enhancing foresight-driven decision-making and system resilience in public health; and (3) clarification of technical boundaries and implementation pathways, yielding a reusable methodology and practical paradigm for AI-enabled global health security governance.

Organizations often struggle to effectively identify and track weak risk signals detected by frontline employees, hindering the development of proactive resilience. This study proposes a human-centered Weak Signal Cultivation Model (WSCM), introducing the novel concept of a “weak signal cultivation field.” By employing a two-dimensional continuous coordinate space, the model dynamically locates and tracks risk signals, mapping them as trajectory nodes characterized by both intensity and growth potential. It further delineates four distinct risk zones, establishing a unified risk lexicon. Integrating dynamic tracking with visual analytics, the WSCM provides a theoretical foundation for AI-augmented risk assessment and delivers an actionable toolkit that bridges frontline sensing with executive decision-making, thereby significantly enhancing organizational capabilities in early risk identification and response.

Existing business process modeling practices separate process flows from business rules, leading to fragmented mental models and impaired comprehension among expert process workers. Method: This study employs a mixed-methods design integrating eye-tracking and concurrent verbal protocol analysis, coupled with cognitive-behavioral coding, to investigate how domain experts perform sensemaking when interpreting integrated process-rule models. Contribution/Results: We identify fine-grained visual search patterns and cognitive bottlenecks that critically affect comprehension efficiency during information foraging and cognitive processing stages. Based on these findings, we propose empirically grounded design principles for personalized cognitive support targeting knowledge workers. The results provide actionable evidence to enhance integrated modeling languages, tool interfaces, and training strategies—advancing business process modeling from syntactic formalism toward cognitive alignment and human-centered design.

This study addresses a novel Lazarus Group campaign targeting cryptocurrency wallets and financial data, focusing on its persistence mechanisms, C2 communication patterns, and data exfiltration tactics. Method: We systematically map the underlying infrastructure and innovatively integrate Tactics, Techniques, and Procedures (TTPs) with multi-source threat intelligence to construct a threat-hunting hypothesis model aligned with the MITRE ATT&CK framework. The methodology combines static and dynamic reverse engineering, IoC correlation mining, and real-time behavioral anomaly detection. Contribution/Results: We derive actionable detection rules and alert-optimization strategies that bridge tactical analysis with strategic risk forecasting. Experimental evaluation demonstrates over a threefold improvement in threat detection speed, significantly enhancing predictive capability against APT behaviors and strengthening defensive resilience.

To address the challenge of detecting Advanced Persistent Threats (APTs) that evade traditional rule-based engines, this paper proposes a lightweight, interpretable predictive analytics framework integrating logistic regression and K-means clustering. Designed for low-resource settings with small-scale security event data (Kaggle dataset, *n* = 2,000), it enables real-time threat detection and response. Methodologically, it is the first to synergistically combine these two models in resource-constrained environments and employs SPSS-based statistical tests to validate feature significance. Compared to baseline rule engines, the framework achieves significantly improved threat alert sensitivity (+23.6%) and reduces average response time by 41%, while preserving high model interpretability. It thus delivers actionable, proactive defense decision support for Security Operations Centers (SOCs).