The rapid development of digital inclusive finance has provided new financing opportunities for small and medium-sized enterprises, but it has also increased the complexity and concealment of credit risks. Traditional static-indicator-based credit assessment is insufficient to capture dynamic risk evolution and provide early warning. In this study, the authors proposed a machine learning credit risk assessment and financial early warning framework for small and medium-sized enterprises in digital inclusive finance. By integrating financial records, online transaction behaviors, digital activity characteristics, and external environmental information, the authors constructed a four-level risk state machine and a multi-period early warning heat matrix; they identified the current risk stage of each enterprise and quantified how many periods in advance the risk can be detected. The authors compared logistic regression, gradient boost decision tree, and Extreme Gradient Boosting, and they introduced an explainable feature importance analysis. Results demonstrated superior risk discrimination and early warning performance.
Social media app databases contain vast user information, serving as critical evidence in legal investigations. This study proposes an innovative forensic framework for analyzing and decrypting social app databases, using WeChat and OpenIM as representative cases. The framework integrates file system analysis, reverse engineering, dynamic debugging, and brute-force attacks to extract and decrypt sensitive data from encrypted databases. Key innovations include code analysis to uncover encryption mechanisms, dynamic debugging to retrieve decryption keys, and brute-force attacks to crack weakly encrypted databases. The proposed framework successfully extracted, decrypted, and visualized critical user data from both WeChat and OpenIM databases. The findings underscore the framework's potential to enhance digital forensic capabilities and ensure secure, effective evidence extraction from social app databases in legal contexts.
Police unmanned aerial vehicles (UAVs) are increasingly used in patrol, rescue, and crime-scene imaging. This study developed a cloud–IoT-based framework for UAV training assessment and management. The framework integrates flight telemetry, task logs, alert records, and instructor feedback to support continuous monitoring and targeted training adjustment. A paired pre–post analysis was conducted using routine training records from 48 trainees. After training, task completion rate, normative operation score, emergency response efficiency, collaborative disposal score, and data-return accuracy increased by 10.6%, 10.8%, 12.4%, 12.8%, and 10.6%, respectively; all reported p < .001. The framework may improve the traceability and standardization of UAV training. Its effects on real-world digital-evidence collection require further validation.
Electronic accounting voucher deposits and cybersecurity audits face a contradiction between resistance to evidence tampering and data privacy protection. Using blockchain technology as a basis, the authors propose a solution framework that addresses both concerns. Under a dual-layer hash mapping architecture, original accounting vouchers are stored off-chain via the InterPlanetary File System, and digital fingerprints are anchored on the immutable ledger. Smart contract engines enforce a multi-signature state machine to ensure accounting voucher traceability from generation to archiving. Paillier homomorphic encryption supports direct computation on ciphertext, enabling continuous consistency checks between accounts and accounting vouchers without decrypting raw data. Experimental results showed that on-chain storage overhead was reduced by 99.08%, audit latency reached 12.4 s under a workload of 5,000 transactions per second, and no raw data were exposed throughout the process. This framework provides forensic-ready support for secure electronic accounting voucher management.
Law enforcement tactical training involves high-pressure scenarios, but traditional methods are limited by difficult scene simulation, incomplete records, insufficient quantitative evaluation, and delayed feedback. Training quality relies heavily on instructors’ experience, leading to inconsistent standards and poor repeatability, especially in multi-agent collaboration and rapid decision-making. This study uses artificial intelligence (AI) to build a full-process closed-loop framework for scenario generation, behavior collection, evaluation, feedback optimization, and retraining. Results confirm AI significantly improves training efficiency, feedback speed, collaboration quality, and capability assessment accuracy. The framework forms a verifiable, replicable training paradigm, and full-process data can serve as standardized digital evidence for law enforcement review and supervision.
