The purpose of this study is to analyse the effectiveness of implementing adaptive AI learning models in decision support systems to optimise the functioning of smart regions. The study provides a detailed examination of the application of machine learning algorithms, deep learning, and reinforcement learning across various sectors, such as urban management, energy resources, and security. The results revealed that the implementation of these models enhances the efficiency of urban system management, reduces costs, and increases the flexibility of decision-making processes. In particular, adaptive models in energy resource management optimise decision-making processes, leading to more rational resource use and substantial cost reductions. In the security field, adaptive AI models show improvements in predicting and preventing incidents, ensuring more reliable and stable system performance. Moreover, the results include the implementation of adaptive models based on programming languages such as TypeScript and JavaScript. The study demonstrated that the use of TypeScript reduces errors and improves system scalability due to strict typing, as shown in the implementation of a reinforcement learning model. Meanwhile, the use of JavaScript enabled the effective adaptation of models to new data through dynamic updates of regression coefficients, leading to improved prediction accuracy.

Speaker identification among identical twins remains a significant challenge in voice-based biometric systems, particularly under emotional variability. Emotions dynamically alter speech characteristics, reducing the effectiveness of conventional identification algorithms. To address this, we propose a hybrid deep learning architecture that integrates gender and emotion classification with speaker Identification, tailored specifically to the complexity of identical twin voices. The system combines Emphasized Channel Attention Propagation and Aggregation in Time Delay Neural Network (ECAPA-TDNN) embeddings for speaker-specific representations, Power Normalized Cepstral Coefficients (PNCC) for noise-robust spectral features, and Maximal Overlap Discrete Wavelet Transform (MODWT) for effective time-frequency denoising. A Radial Basis Function Neural Network (RBFNN) is employed to refine and fuse feature vectors, enhancing the discrimination of emotion-related cues. An attention mechanism further emphasizes emotionally salient patterns, followed by a Multi-Layer Perceptron (MLP) for final classification. The model is evaluated on speech datasets from RAVDESS, Google Research, and a proprietary corpus of identical twin voices. Results demonstrate significant improvements in speaker and emotion recognition accuracy, especially under low signal-to-noise ratio (SNR) conditions, outperforming traditional Mel Cepstral-based methods. The proposed system’s integration of robust audio fingerprinting, feature refinement, and attention-guided.

This paper presents a conceptual framework for enhancing knowledge management (KM) processes in virtual organizations through the integration of mobile agents. With the growing digitization of workplaces and the proliferation of distributed teams, managing and leveraging knowledge efficiently has become critical. Mobile agents offer promising features such as autonomy, adaptability, and mobility, making them suitable for dynamic knowledge environments. The paper outlines the architecture of a multi-agent system for KM and discusses its potential impact on organizational performance. Emphasis is placed on the role of intelligent agents in collecting, filtering, and disseminating relevant knowledge across virtual settings. The proposed model aims to support decision-making, reduce information overload, and facilitate knowledge sharing among members of decentralized organizations.

The environmental performance and service quality of liquefied petroleum gas (LPG) vehicles were evaluated through a dual-phase analytical approach. In the first phase, exhaust emissions from LPG and petrol-powered vehicles were quantified using the CAPELEC 3010 gas analyzer, with concentrations of carbon monoxide (CO), carbon dioxide (CO$_2$), nitrogen oxides (NOx), and hydrocarbons being measured. The results demonstrated that LPG vehicles emitted significantly lower CO levels (0.09% on average) compared to petrol vehicles (0.18%), with corrected CO values also reduced (0.08% vs. 0.19%). These findings reinforce the environmental advantages of LPG as a cleaner fuel alternative. In the second phase, the SERVQUAL model was employed to assess user perceptions of service quality, focusing on five dimensions: reliability, responsiveness, assurance, empathy, and overall service quality. A negative overall SERVQUAL gap (-0.806) was identified, with the most pronounced discrepancies observed in reliability (-1.061) and responsiveness (-0.933), indicating unmet expectations in key service aspects. Despite these gaps, LPG vehicles were perceived as cost-effective and environmentally sustainable. The findings underscore the necessity for technical refinements in LPG vehicle systems and improvements in service infrastructure to enhance user satisfaction. The insights derived from this study offer valuable guidance for policymakers and industry stakeholders seeking to promote LPG as a viable component of sustainable transportation strategies.

Sentiment analysis in legal documents presents significant challenges due to the intricate structure, domain-specific terminology, and strong contextual dependencies inherent in legal texts. In this study, a novel hybrid framework is proposed, integrating Graph Attention Networks (GATs) with domain-specific embeddings, i.e., Legal Bidirectional Encoder Representations from Transformers (LegalBERT) and an aspect-oriented sentiment classification approach to improve both predictive accuracy and interpretability. Unlike conventional deep learning models, the proposed method explicitly captures hierarchical relationships within legal texts through GATs while leveraging LegalBERT to enhance domain-specific semantic representation. Additionally, auxiliary features, including positional information and topic relevance, were incorporated to refine sentiment predictions. A comprehensive evaluation conducted on diverse legal datasets demonstrates that the proposed model achieves state-of-the-art performance, attaining an accuracy of 93.1% and surpassing existing benchmarks by a significant margin. Model interpretability was further enhanced through SHapley Additive exPlanations (SHAP) and Legal Context Attribution Score (LCAS) techniques, which provide transparency into decision-making processes. An ablation study confirms the critical contribution of each model component, while scalability experiments validate the model’s efficiency across datasets ranging from 10,000 to 200,000 sentences. Despite increased computational demands, strong robustness and scalability are exhibited, making this framework suitable for large-scale legal applications. Future research will focus on multilingual adaptation, computational optimization, and broader applications within the field of legal analytics.

The transformative potential of artificial intelligence (AI) and machine learning (ML) in healthcare has been increasingly recognized, particularly in medical image analysis and predictive modeling of patient outcomes. In this study, a novel convolutional neural network (CNN) architecture incorporating customized skip connections was introduced to enhance feature extraction and accelerate convergence during medical image classification. This model demonstrated superior performance compared with conventional architectures such as Residual Network with 50 layers (ResNet-50) and Visual Geometry Group with 16 layers (VGG16), achieving an accuracy of 96.5% along with improved precision, recall, and area under the receiver operating characteristic curve (AUC-ROC). In parallel, patient readmission risks were predicted using an optimized random forest algorithm, which, after hyperparameter tuning, attained a robust AUC-ROC value of 0.91, thereby underscoring its stability and predictive reliability. The integration of these approaches highlights the ability of AI and ML systems to deliver more accurate diagnoses, anticipate potential health risks, and recommend personalized treatment strategies, ultimately enabling faster and more precise clinical decision-making. Despite these advancements, challenges persist regarding data privacy, interpretability of AI-driven decisions, and the ethical use of patient information. Addressing these limitations will be critical for the broader adoption of AI-enabled healthcare systems. The findings of this study reinforce the role of advanced AI and ML frameworks in improving healthcare delivery, optimizing the use of limited resources, and reducing operational costs, thereby supporting more effective patient care.

The restoration of blurred images remains a critical challenge in computational image processing, necessitating advanced methodologies capable of reconstructing fine details while mitigating structural degradation. In this study, an innovative image restoration framework was introduced, employing Complex Interval Pythagorean Fuzzy Sets (CIPFSs) integrated with mathematically structured transformations to achieve enhanced deblurring performance. The proposed methodology initiates with the geometric correction of pixel-level distortions induced by blurring. A key innovation lies in the incorporation of CIPFS-based entropy, which is synergistically combined with local statistical energy to enable robust blur estimation and adaptive correction. Unlike traditional fuzzy logic-based approaches, CIPFS facilitates a more expressive modeling of uncertainty by leveraging complex interval-valued membership functions, thereby enabling nuanced differentiation of blur intensity across image regions. A fuzzy inference mechanism was utilized to guide the refinement process, ensuring that localized corrections are adaptively applied to degraded regions while leaving undistorted areas unaffected. To preserve edge integrity, a geometric step function was applied to reinforce structural boundaries and suppress over-smoothing artifacts. In the final restoration phase, structural consistency is enforced through normalization and regularization techniques to ensure coherence with the original image context. Experimental validations demonstrate that the proposed model delivers superior image clarity, improved edge sharpness, and reduced visual artifacts compared to state-of-the-art deblurring methods. Enhanced robustness against varying blur patterns and noise intensities was also confirmed, indicating strong generalization potential. By unifying the expressive power of CIPFS with analytically driven restoration strategies, this approach contributes a significant advancement to the domain of image deblurring and restoration under uncertainty.

In recent decades, the strategic placement of capacitors for compensating inductive reactive power has been extensively investigated by network operators and researchers globally, owing to its profound impact on minimizing power losses, improving voltage regulation, and enhancing overall voltage stability. The installation of shunt capacitors has been demonstrated to significantly improve the efficiency and performance of power systems by regulating voltage levels at load points, as well as at distribution and transmission system buses. This approach not only reduces inductive reactive power but also corrects the system’s power factor, thereby optimizing energy utilization. In this study, the optimal sizing and placement of capacitor banks within a specific section of the Duhok city distribution network were systematically analyzed. The Electrical Transient Analyzer Program (ETAP) software was employed to simulate and evaluate power losses and voltage drops both before and after capacitor installation. The findings reveal a marked improvement in the voltage profile across the network, accompanied by a substantial reduction in power losses. These results underscore the critical role of capacitor banks in enhancing the operational efficiency of distribution networks, providing a robust framework for future implementations in similar systems. The methodology and outcomes presented herein offer valuable insights for network operators seeking to optimize power system performance through reactive power compensation.

The domination of cyberspace technologies in inter-human communications is obvious because of their ‎ultra-rapidness and enormous data capacity. Human-intensive ‎use of cyberspace increased the magnitude of streamed data through its nodes, created by two sources: human users and AI. While humans can control their generated data, ‎it proves impossible to control AI due to its super intelligence along with their self-developing ‎abilities, enabling it to produce unlimited volumes of data. It is known that cyberspace depends on physical infrastructure, which is inherently limited. Despite investments to expand capacity, overloading this infrastructure with unlimited data creates critical functionality issues. Additionally, the presence of uncontrollable AI elements leads to unpredictable outcomes. Ultimately, this results in AI dominating cyberspace, a phenomenon known as cyber singularity.

The ultimate consequences of AI cyber singularity motivated the study to recall a similar phenomenon in astrophysics: gravitational singularity. Using general relativity theory, the ‎research analyses the dilemma of data overload in cyberspace and its effects, drawing parallels ‎between outer space and cyberspace‎. It aims to illustrate AI's acquisition of cyber singularity according to astrophysics laws on gravitational singularity, providing an innovative perspective for scientists and scholars studying cyberspace.

This article aims to discuss the evolution over the centuries of the role and social position of those mastering the technologies of their time. We suggest that the Industrial Revolution, the rationalization of technical and managerial processes, then the rise of IT, the ascent of cryptocurrencies and finally the emergence of the neoliberal state have lifted a fringe of these individuals to the top of the social hierarchy. Among the “technology masters”, we distinguish three families: those who remain at the service of the State and the established order, those who have exploited, consciously or not, the withdrawal of the neoliberal State to offer services and innovations formerly assumed by the public sector, and finally those who have consciously taken advantage of this same withdrawal and the recognition they enjoy in society to propose other models (free software, open source, crypto anarchism, technical alternatives, etc.).