This study provides a comprehensive analysis of cognitive password systems as a secure and user-friendly alternative to traditional authentication mechanisms. Cognitive passwords leverage human memory, behavior, and perception to enhance usability while mitigating common security challenges such as poor memorability, password reuse, and susceptibility to attacks. The study systematically reviews various models, including graphical passwords, cognitive biometrics, and cognitive one-time passwords (OTPs), highlighting their strengths and limitations. To ensure a transparent and rigorous review, we employed a structured methodology comprising a multi-database literature search, clearly defined inclusion and exclusion criteria, and a thematic synthesis of the collected studies. Our findings indicate that cognitive password systems offer significant improvements in user experience and security but face critical challenges, including accessibility for individuals with cognitive or physical impairments, privacy concerns, vulnerability to social engineering, and scalability limitations. Furthermore, artificial intelligence (AI) emerges as a key enabler for enhancing personalization, adaptive authentication, and real-time security risk assessment. The study underscores the necessity of integrating AI thoughtfully to maximize the benefits of cognitive passwords. Overall, this research demonstrates the transformative potential of cognitive password systems in cybersecurity, emphasizing that addressing usability, privacy, and scalability challenges is essential for their practical adoption. The findings provide actionable insights for system designers, policymakers, and researchers aiming to advance secure and user-centered authentication frameworks.

• View PDF

Phishing attacks pose significant risks in the digital landscape, resulting in financial losses and sensitive information breaches. Traditional detection methods often struggle to keep pace with evolving threats, compromising their effectiveness. This study addresses these limitations by developing a robust detection system using a hybrid machine learning approach. We combine random forest, gradient boosting, and logistic regression algorithms to enhance phishing detection accuracy. A labeled dataset of URLs from Kaggle is utilized, with robust feature engineering extracting key attributes for model training. Following the CRISP-DM framework and leveraging Object-Oriented Programming principles, we develop a model that achieves strong performance metrics. The model's accuracy stands at 84%, with precision, recall, and F1-score values of 85%, 86%, and 84%, respectively. Notably, the model demonstrates excellent ability to differentiate between phishing and legitimate URLs, with an ROC AUC score of 91%. These results confirm the model's potential as a reliable phishing detection tool, capable of identifying phishing URLs effectively while minimizing false positives. Our research contributes to the development of more effective phishing detection strategies, ultimately safeguarding users and organizations from economic and reputational harm. By leveraging machine learning, we can develop more robust cybersecurity systems. Our proposed model can be seamlessly integrated into existing security frameworks to improve the detection of phishing threats.

• View PDF

Wireless networks generate large volumes of heterogeneous data from network elements, user equipment, and management systems, posing significant challenges for effective network monitoring, fault management, and resource optimization. Traditional rule-based or data-driven approaches often lack unified knowledge representation and reasoning capability, limiting their scalability and interpretability. To address these challenges, this paper proposes a knowledge-graph-based framework for wireless network knowledge construction, management, and application. The proposed framework integrates multi-source network data through ontology-driven modeling and rule-based semantic mapping, enabling structured representation of network entities, events, and their relationships. An event-driven incremental update mechanism is introduced to efficiently maintain the knowledge graph in dynamic network environments without full reconstruction. Furthermore, a lightweight reasoning mechanism is employed to infer implicit network states and support intelligent network management decisions. The framework is designed to balance expressiveness and computational efficiency, making it suitable for large-scale wireless networks. To quantitatively evaluate the effectiveness of the proposed approach, extensive experiments are conducted under different network scales. The experimental results demonstrate that the proposed framework consistently outperforms traditional rule-based methods in terms of fault localization accuracy and resource utilization efficiency, while exhibiting lower query latency and better scalability as the network size increases. The results indicate that the proposed knowledge-graph-based framework provides an effective and scalable solution for intelligent wireless network management, with potential applicability to fault detection, resource optimization, and network security analysis.

• View PDF

Star-connected computer networks are explored using the example of terahertz modeling. Existence conditions for the complete synchronization between constituent lasers are found numerically. Numerical simulations were conducted using the Matrix Laboratory (MATLAB) software to solve Delay Differential Equations. Extensive numerical simulations with different initial states confirm that high-quality, near-perfect complete synchronization between terahertz lasers occurs. As in the real world, parameters can differ; we simulated the star-connected computer network model with parameter mismatches of 3–5%. Still, we have obtained close to 100% of correlation between the dynamics of terahertz lasers. Synchronization is important in chaos-based communication. It is underlined that chaos-based communication security between computer networks can offer an additional layer of security to the traditional cryptography based on the Rivest, Shamir, and Adleman (RSA) algorithm. This algorithm uses the mathematical challenge of factoring very large prime numbers. The extra layer of security is of immense importance in light of the exponential increase in central processing units (CPUs) in computers. This is especially true in light of the quantum processing unit (QPU), which is the core processor of a quantum computer, using qubits in superposition and entanglement to perform complex, parallel calculations far beyond classical CPUs. Unlike traditional binary CPUs, QPUs excel at optimization, cryptography, and artificial intelligence tasks. 

• View PDF

The rise of short videos and live streaming has transformed digital governance, enabling governments to engage with citizens in real time. This paper explores the integration of artificial intelligence (AI) in government live streaming, focusing on its role in enhancing public opinion monitoring and sentiment analysis. We analyze current practices and introduce innovative solutions that leverage AI for improved content dissemination and audience engagement. Using the SO-PMI (Semantic Orientation Pointwise Mutual Information) algorithm, we conduct sentiment analysis on audience comments to capture emotional nuances. Additionally, large language models are employed for efficient live data processing, enabling real-time comment handling and generating responsive communication tailored to audience sentiment. The methodology includes robust data collection from social media platforms, employing API connections to retrieve comments during broadcasts. Extensive preprocessing involves filtering irrelevant content, normalization, tokenization, and lemmatization. Sentiment analysis categorizes comments as positive, negative, or neutral, while real-time monitoring allows presenters to adapt their messaging based on audience sentiment. Our experimental results demonstrate the effectiveness of AI-driven methods in managing real-time comments, providing accurate emotional analyses, and facilitating prompt responses. This study illustrates AI’s potential to engage citizens more effectively and enhance governmental communication strategies, ultimately fostering transparency and responsiveness in the digital age.

• View PDF

In today's rapidly changing business environment, organizations and supply chains face increasing socio-economic challenges. To address these challenges, they need new sources of growth and development. One of these sources is lost profits or profits that could be recouped by maximizing the benefits provided by the external environment. To reduce lost profits, organizations and supply chain need intellectual support based on the knowledge and experience of famous managers from past and current generations. Memoirs of these managers presented in the form of software products can be used to share their insights and best practices with current and future managers. The purpose of this article is to develop recommendations for creating digital memoirs of famous managers to design a digital twin for managing organizations and supply chains allowing them to minimize lost profits. To achieve this purpose, terminological analysis, descriptive and faceted methods of qualitative research of non-physical management objects were used. These methods have made it possible to identify, formalize, structure, combine, and digitize objects of this type. This article develops a classification and substantiates the structure of digital systems formed in organizations and supply chains. It also creates a template and form of digital memories of famous managers for use in the digital twin of management of these organizations and supply chains. The results obtained allow us to create an artificial intelligence that operates with non-physical management objects and еру digital twin to manage the organization and supply chain. This ensures the development of management decisions with minimal loss of profit.

• View PDF

The rapid expansion of social media platforms has intensified the spread of negative sentiments, misinformation, and hostile digital interactions, producing measurable consequences for public discourse, institutional trust, and societal stability. Existing models often overlook the cognitive and network-structural mechanisms that drive negative sentiment cascades, limiting capacity for early detection and effective intervention. The proposed study introduces the Expression Capacity-based Negative Sentiment (ECNS) Mitigation framework that integrates graph-theoretic analysis with psychological modeling to better understand and mitigate sentiment propagation. The ECNS architecture consists of five interconnected models dedicated to data ingestion, influencer identification, cascade monitoring, intervention control, and final sentiment-state generation. The operational core relies on two algorithms: the first identifies high-impact influencer nodes using community-level analysis and expression-capacity thresholds, while the second monitors negativity diffusion, evaluates sequential comment behavior, and applies targeted mitigation based on capacity depletion and sentiment intensity. This design enables proactive control of sentiment cascades rather than reactive moderation. The framework was evaluated across three heterogeneous datasets ResearchGate, Zhihu, and Sentiment140 to reflect diverse interaction patterns and topical domains. Comparative performance against three leading models such as EANN, HMCan, and AOAN demonstrates that ECNS provides more accurate influencer detection, stronger cascade containment, and more effective sentiment reduction. Overall, ECNS achieved improvements ranging from 5.8% to 11.4% points across key evaluation metrics, confirming its capacity to suppress negative sentiment propagation with significantly higher reliability.

• View PDF

In the realm of secure digital communication, balancing payload capacity with perceptual imperceptibility remains a critical challenge, particularly within lossy compression standards like JPEG. This paper proposes a robust steganography framework that addresses this trade-off by integrating adaptive Discrete Cosine Transform (DCT) modification with an edge-based selection criterion. Unlike uniform embedding approaches that indiscriminately treat all image regions, the proposed method employs Canny edge detection to categorize blocks based on texture complexity. This strategy leverages the masking properties of the Human Visual System (HVS), allocating higher payload capacities to complex, edge-rich regions where modifications are statistically less detectable. To further enhance embedding efficiency, a decimal-to-ternary (base-3) coding scheme is introduced to optimize the utilization of DCT coefficients. This mechanism is coupled with a modulo optimization search within a constrained range of , which minimizes the magnitude of necessary modifications compared to traditional binary embedding. Experimental evaluations on standard datasets, including USC-SIPI, indicate that the method maintains a Peak Signal-to-Noise Ratio (PSNR) significantly higher than traditional methods, ranging from 48 dB to 62 dB depending on the embedding rate. Furthermore, quantitative analysis demonstrates a 58.5% improvement in embedding efficiency over standard binary techniques. Consequently, this adaptive strategy offers a superior trade-off between high-capacity data hiding and resistance to statistical steganalysis compared to standard LSB and non-adaptive DCT methods.

• View PDF

Contemporary hybrid threats employ coordinated campaigns across information, cyber, and physical domains, maintaining plausible deniability while exploiting institutional vulnerabilities. This review conducts a scoping analysis following the PRISMA ScR framework (Preferred Reporting Items for Systematic Reviews and Meta-Analyses—Extension for Scoping Reviews) to evaluate Open Source Intelligence (OSINT), Social Media Intelligence (SOCMINT), and Natural Language Processing (NLP) capabilities relevant to hybrid threat detection. We systematically assess these technologies against a 12-point operational requirements framework derived from documented Russian and Chinese military OSINT methodologies and influence operation tradecraft. The analysis incorporates high-Technology Readiness Level (TRL) European initiatives to ground capability assessments in operational experience. While individual analytical disciplines are technically advanced, current defensive systems remain siloed and lack the cross-domain reasoning necessary to correlate technical cyber indicators with coordinated narrative manipulation. Requirements for cross-platform correlation and adversarial adaptation show only prototype stage coverage. Our findings reveal a persistent “semantic gap”: defensive systems collect extensive data but lack integrated semantic reasoning across domains and languages. To address this, we examine ontology-based approaches as architectural solutions, positioning the ‘Hybrid Information Psychological Societal Threats handling system’ (HIPSTer) framework as an illustrative case. HIPSTer specifically targets the multilingual nature of hybrid threats—particularly in Russian and Chinese contexts—achieving TRL-4 validation through high-efficiency semantic vectors and formal reasoning across diverse language benchmarks. Finally, the review analyzes how European regulations—including the General Data Protection Regulation (GDPR), AI Act, and Network and Information Systems Directive 2 (NIS2)—shape operational architectures through compliance by design imperatives. We conclude by outlining a prioritized research agenda to advance European hybrid threat detection toward operational maturity.

• View PDF

Dew Computing (DC) has recently emerged as a complementary computing paradigm that extends cloud, fog, and edge computing by enabling autonomous, local-first computation directly at end-user devices. Unlike traditional distributed models that rely on centralized or near-edge infrastructures, Dew Computing emphasizes offline-capable, low-latency, and resilient processing at the extreme edge, while maintaining synchronization with fog and cloud layers when connectivity is available. This paper presents a systematic and comprehensive review of Dew Computing based on a structured literature analysis, covering its conceptual foundations, architectural models, and operational mechanisms. The study analyzes key Dew-based architectures, including cloud–dew and hybrid edge frameworks, highlighting their role in reducing latency, improving fault tolerance, enhancing energy efficiency, and supporting privacy-preserving local processing. In contrast to existing surveys, this work provides a critical synthesis of current approaches by identifying their strengths, limitations, and deployment trade-offs across different application scenarios. Furthermore, the paper examines major application domains such as Internet of Things (IoT), smart healthcare, smart agriculture, and cyber-physical systems, where Dew Computing demonstrates advantages in real-time responsiveness and operational resilience. Security and privacy challenges are also analyzed, focusing on recent solutions such as blockchain-based trust management, federated learning, lightweight cryptographic protocols, and AI-driven intrusion detection, while highlighting unresolved issues related to scalability and resource constraints. Unlike prior works, non-computing interpretations such as meteorological dew-point modeling are excluded or clearly distinguished to avoid conceptual ambiguity. Finally, the survey identifies open research challenges, adoption barriers, and future research directions, positioning Dew Computing as a key enabler for decentralized, user-centric, and resilient next-generation computing systems.

• View PDF