Internet of Things deployments face escalating security threats, yet systematic methods for auditing the defensive posture of IoT device networks remain underdeveloped. Current intrusion detection evaluations focus on algorithmic accuracy while neglecting operational requirements—computational efficiency, reproducibility, and interpretable risk assessment—that security audits demand. This paper introduces a reproducible security audit framework for IoT device networks, demonstrated through systematic evaluation of four machine learning models (Random Forest, LightGBM, XGBoost, Logistic Regression) on the TON_IoT dataset containing nine attack categories targeting smart environments. Our audit methodology enforces strict feature hygiene by excluding identity-revealing attributes, benchmarks both threat detection capability and computational cost, and provides complete reproducibility artifacts including preprocessing pipelines and trained models. The framework evaluates security posture through dual lenses: binary classification (distinguishing compromised from legitimate traffic) and multiclass classification (attributing threats to specific attack types). Binary audit results show ensemble models achieve 99.8–99.9% accuracy with perfect ROC-AUC (100%) and sub-15 ms inference latency per 1000 flows, confirming reliable attack detection. Multiclass auditing reveals more nuanced findings: while overall accuracy reaches 99.0% with macro-F1 near 97%, rare attack types expose critical blind spots—man-in-the-middle threats achieve only 78% F1 despite representing serious security risks. LightGBM provides optimal audit performance, balancing 99.93% detection accuracy with 2.76 MB deployment footprint. We translate audit findings into actionable security recommendations (network segmentation, rate-limiting, TLS metadata collection) and compare against twenty published studies, demonstrating that our framework achieves competitive detection rates while uniquely delivering the transparency, efficiency metrics, and reproducibility required for credible security assessment of production IoT networks.

Educational equity and access to quality learning opportunities represent fundamental pillars of sustainable societal development, directly aligned with the United Nations Sustainable Development Goal 4 (Quality Education). Student retention remains a critical challenge in higher education, with early disengagement strongly predicting eventual failure and limiting opportunities for social mobility. While machine learning models have demonstrated impressive predictive accuracy for identifying at-risk students, most systems prioritize performance metrics over practical deployment constraints, creating a gap between research demonstrations and real-world impact for social good. We present an accountable and interpretable decision support system that balances three competing objectives essential for responsible AI deployment: ultra-early prediction timing (day 14 of semester), manageable instructor workload (flagging 15% of students), and model transparency (multiple explanation mechanisms). Using the Open University Learning Analytics Dataset (OULAD) containing 22,437 students across seven modules, we develop predictive models from activity patterns, assessment performance, and demographics observable within two weeks. We compare threshold-based rules, logistic regression (interpretable linear modeling), and gradient boosting (ensemble modeling) using temporal validation where early course presentations train models tested on later cohorts. Results show gradient boosting achieves AUC (Area Under the ROC Curve, measuring discrimination ability) of 0.789 and average precision of 0.722, with logistic regression performing nearly identically (AUC 0.783, AP 0.713), revealing that linear modeling captures most predictive signal and makes interpretability essentially free. At our recommended threshold of 0.607, the predictive model flags 15% of students with 84% precision and 35% recall, creating actionable alert lists instructors can manage within normal teaching duties while maintaining accountability for false positives. Calibration analysis confirms that predicted probabilities match observed failure rates, ensuring trustworthy risk estimates. Feature importance modeling reveals that assessment completion and activity patterns dominate demographic factors, providing transparent evidence that behavioral engagement matters more than student background. We implement a complete decision support system generating instructor reports, explainable natural language justifications for each alert, and personalized intervention templates. Our contribution advances responsible AI for social good by demonstrating that interpretable predictive modeling can support equitable educational outcomes when designed with explicit attention to timing, workload, and transparency—core principles of accountable artificial intelligence.

The S-box is a key component of modern ciphers, determining the quality and performance of the cryptographic algorithms in which it is applied. Many constructions for synthesizing high-quality S-boxes have been established, and those based on Galois fields theory—for example, the Nyberg construction applied in the AES cryptographic algorithm—are particularly important. An integral component of the Nyberg construction is the affine transformation, which is used to improve the avalanche and correlation properties of the S-box. In this paper, a new approach is adopted for synthesizing affine transformations for S-boxes based on the quaternary matrices over the Galois field GF(4). We describe four basic structures that serve as the foundation for synthesizing a complete class of 648 affine transformation matrices of order n = 3 and a class of 7776 matrices of order n = 4 and introduce a recurrent structure to facilitate the synthesis of matrices for higher orders. Using these matrices in combination with the Nyberg construction, it is possible to construct bijective S-boxes that outperform the original Nyberg construction and many other known S-boxes in terms of strict avalanche criterion (SAC) and bit independence criterion strict avalanche criterion (BIC SAC) values, while maintaining a maximal level of nonlinearity and good cryptographic properties. We also propose modified GF(4) affine transformations that can be applied to specialized S-boxes which already satisfy the SAC for both component Boolean and 4-functions, as well as the criterion of minimal correlation between input and output, allowing us to enhance their nonlinearity to the value of N_f = 96. We integrate the synthesized S-boxes into the AES algorithm and evaluate their practical performance. The encryption outputs successfully pass the NIST statistical test suite in 96 out of 100 cases, outperforming both the original AES S-box and other reference constructions, confirming the practical strength of the proposed method.

Updating a TinyML intrusion detector can change its decisions even when an authentic artifact is correctly installed: preprocessing, numerical representation, learned parameters, and the operating threshold jointly define the deployed function. We formalize this release as a pipeline contract and distinguish authenticated installation, decision agreement with a named reference, and detection quality. Signed pipeline packages and journaled release selection implement the contract on an ESP32-S3 and two ESP32-C3 boards. Evaluation combines chronological TON_IoT scenarios, a separate CICIoT2023 release experiment, host–device numerical checks, and installation experiments. Retaining an earlier threshold changed 15% of the floating-point MLP’s decisions. Export rules exact in real arithmetic still changed implemented decisions (9.3% for a threshold-folded logistic-regression export in one scenario), and targeted boundary inputs revealed host–device discrepancies absent on ordinary traffic. Alternative quantization improved the tested operating point while changing the reference integer release. Signed packages of 448–1588 bytes, up to 800 times smaller than the application image, reduced on-device installation processing about 40-fold on S3 and 14–16-fold on C3 relative to raw and compressed full-image delivery. During package installation, 185 of 188 rail-confirmed physical supply interruptions returned an expected release, with no unexpected or mixed release observed. These results support treating the complete pipeline as the unit of deployment and validating its decisions against an explicit reference.

The article is dedicated to assessing the impact of digitalization on educational practices and the professional activities of teachers. The aim of the study was to analyze primary data collected from the teaching community to identify the current level of digital competencies, the nature of the use of digital tools and artificial intelligence (AI) technologies, as well as their influence on the educational process and student outcomes. The research methodology included a quantitative and qualitative survey of 832 teachers from various educational institutions, conducted through an online questionnaire. The data collection covered demographic characteristics of the respondents, the frequency of use of digital tools (MS Office, Kahoot) and AI (ChatGPT, Gemini), the use of digital technologies in the classroom (interactive testing, AR), subjective assessment of their impact on student motivation and performance, as well as the need for professional development. The results of the study revealed active integration of digital tools, with MS Office being the most commonly used and growing popularity of AI tools. Teachers noted a significant increase in student engagement (84.6%) and improvement in content comprehension due to digitalization. However, despite the overall positive trend, 4.4% of respondents indicated negative aspects, such as decreased comprehension or increased time costs. A high level of interest in mastering interactive tasks, AR technologies, and AI wasidentified among educators, confirming the demand for continued professional development. The obtained data provide an empirical basis for the development of targeted training programs and the optimization ofstrategies for integrating digital technologies into the educational environment.