Computational intelligence for emotion recognition in autism spectrum disorder: a systematic review of signal-based modeling, simulation, and clinical potential

Sales compensation has long resisted systematic optimization despite its central role in driving organizational performance. Traditional approaches rooted in historical benchmarks and managerial intuition struggle with the mounting complexity of modern B2B sales environments. Machine learning now promises to revolutionize incentive design by processing vast datasets to identify patterns invisible to human analysts and generate recommendations that supposedly balance competing objectives. Yet amid the enthusiasm, a troubling question persists: does the technology actually deliver? This review critically examines what we know and more importantly, what we don't about AI-powered sales incentive systems. Drawing on empirical studies, theoretical frameworks, and implementation experiences across behavioral economics, organizational psychology, and computational intelligence, we find a substantial gap between predictive capability and prescriptive value. While algorithms can forecast performance with reasonable accuracy, evidence that AI-optimized compensation improves business outcomes remains surprisingly thin. More concerning, we identify serious risks around algorithmic bias, unintended behavioral consequences, and over-optimization that organizations have barely begun to address. The field stands at a critical juncture where sober assessment matters more than technological optimism.

In the rapidly evolving fields of Artificial Intelligence (AI) and High Performance Computing (HPC), benchmarking is a critical tool for optimizing system performance.

In this paper, a permanent-magnet-assisted synchronous reluctance motor (SYNRM) coupled with a newly built QDBC and a voltage-fed inverter (VFI) for a standalone PV water pumping system is suggested. Because power supply oscillations can result in short-term disruptions that affect drive performance in industrial applications involving these motors, a robust smooth control system is required to guarantee high efficiency and uninterrupted operation. According to the suggested architecture, a newly built quadratic boost regulator with a very high voltage gain, called a quadruple-diode boost converter (QDBC), is used to first elevate PV voltage to high levels. Additionally, to optimize the power output of the solar PV module, the perturbation and observation highest power point tracking approach (P&O) is implemented. To provide smooth synchronous motor starting, field-oriented control (FOC) of a voltage-fed inverter (VFI) is combined with hysteresis current control of the QDBC. The optimization algorithms discussed in this paper aim to enhance the efficiency of the SYNRM, particularly in operating a synchronous motor powered by variable energy sources such as solar PV. These algorithms function within a cybernetic system designed for water pumping, incorporating feedback loops and computational intelligence for improved performance. Afterward, the three-phase permanent-magnet synchronous motor that drives the mechanical load is fed by the resulting voltage via a voltage source inverter. Furthermore, a thorough hysteresis current control method implementation of the QDBC was suggested in order to attain optimal efficiency in both devices, which is crucial when off-grids are present. Even when the DC-link voltage dropped by up to 10% of the rated voltage, the suggested method was shown to maintain the required reference torque and rated speed. To verify the efficacy of the suggested method, a simulation setup according to the MATLAB 2022b/Simulink environment was employed. To gather and analyze the data, multiple scenarios with varying operating conditions and irradiance levels were taken into consideration. Finally, a working prototype was constructed in order to validate the mathematical analysis and simulation findings of the suggested framework, which includes a 1 kW motor, current sensor, voltage sensor, QDBC, and VCS inverter.

The convergence of the textile industry and electronics presents new frontiers in functional apparel. This research details a system for personalized athletic guidance, rooted in advanced textile processing and materials science. The foundation is a smart garment fabricated from a blend of synthetic fibers using industrial knitting techniques. These methods are also applicable to natural fibers such as cotton or wool for enhanced comfort. The core innovation is a novel piezoresistive yarn, developed by dip-coating a base polyester yarn with poly (3,4-ethylenedioxythiophene):polystyrene sulfonate (PEDOT:PSS), a conductive polymer. This process transforms conventional fiber into a motion-sensing element. A lightweight machine learning model is embedded within the garment’s microcontroller, enabling on-device analysis of data from the yarn sensors. This approach avoids energy-intensive cloud computing and represents a step toward sustainable development in electronic textiles by minimizing the system’s overall energy footprint. The system accurately interprets complex movements, providing real-time feedback. This work demonstrates the integration of computational intelligence directly into fiber products, showcasing a scalable manufacturing pathway for a new generation of interactive textiles and moving beyond traditional applications of materials such as leather for wearable enclosures.

The article provides a comprehensive systemic analysis of the interrelationship between the development of information technologies and the transformation of military unmanned aerial vehicles (UAVs). It is demonstrated that modern warfare, particularly the full-scale aggression against Ukraine, has evidenced the shift of UAVs from an auxiliary means to a leading instrument for reconnaissance, fire adjustment, strike operations, and more. The significance of this research lies in its uncovering of the key role of information technologies in modern warfare, identifying and analyzing the critical technologies that provide a strategic and tactical advantage. The study covers the historical evolution of UAVs and details the synergistic impact of five key IT innovations on their functionality: Artificial Intelligence, sensor fusion, cloud and edge computing, the Internet of Things, and network architectures for swarms. The following research methods (at the theoretical level) were utilized in the article: analysis of scientific literature, systemic analysis, and comparative analysis. Based on the analysis, key areas for further research are identified. This is about the need to overcome existing technological challenges and achieve maximum autonomy and resilience of systems in difficult combat conditions: ‒ creating hybrid drone swarm architectures that combine the advantages of various communication technologies to ensure maximum resilience, in particular artificial intelligence, swarm intelligence, self-learning algorithms, etc.; ‒ developing AI algorithms that can explain their decisions, in order to ensure "human control" (operator control), that is, to best integrate human judgment into autonomous systems without reducing their effectiveness; ‒ developing and implementing the latest cryptographic methods for protecting communication channels for data transmission to ensure the reliability, accuracy and timeliness of data transmission.

The subject matter of the study is the process of selecting the head node of a cluster in wireless sensor networks (WSNs) using intelligent approaches that can adapt to changing environmental conditions. WSNs consist of a large number of sensor nodes with that collect, process and transmit data. Effective clustering is one of the main mechanisms for optimizing the operation of WSNs, as it allows reducing energy consumption, increasing network reliability and scalability. The goal of the study is to analyze the features of using modern computational intelligence tools and methods to increase the efficiency of the sensor node clustering process, which allow taking into account a variety of factors when making decisions about cluster formation and selecting head nodes. Traditional clustering algorithms are not always able to adapt to changes in network parameters, especially in the presence of heterogeneous nodes or changes in topology. In this regard, methods based on computational intelligence, in particular genetic algorithms, neural networks, fuzzy logic, as well as hybrid approaches, are becoming increasingly relevant. These methods allow taking into account a number of parameters when forming clusters and selecting cluster heads. Tasks of the study are analysis of existing approaches to clustering in BSM; development of a clustering fuzzy inference system; construction of a rule base for making optimal decisions; experimental verification of the proposed system. Methods of the study are tools of computational intelligence, in particular neural network learning, genetic optimization and fuzzy control, as well as computer modeling. The article analyzes the advantages of using each of the existing approaches. Results are: a structure of the fuzzy inference system was developed, input and output variables were determined, a database of fuzzy rules and membership functions was formed. The operation of the fuzzy system was simulated in the MATLAB environment. The developed system was also optimized and its operation validated. Conclusions: the use of hybrid intelligent approaches has significant advantages for solving clustering problems in BSM, which may indicate the prospects for further development of systems capable of functioning effectively in conditions of limited resources and high environmental complexity.

Aiming at the problems of low efficiency in traditional paper-and-pencil homework grading and the difficulty of digitizing process data, this study proposes a solution to automate the collection and grading of free-format handwritten homework. To this end, a "cloud-edge-end" collaborative system architecture based on computer vision is proposed. The system first uses cascaded image preprocessing and deep learning semantic segmentation models to accurately analyze the homework layout and locate the question areas. It then employs handwriting recognition models trained with domain adaptation and formula recognition models with context perception of the question stem to complete the structural extraction of the answer content. Finally, combining rule matching and semantic similarity calculation, it achieves intelligent grading of both objective and subjective questions. Experimental results show that on the self-built real-world dataset, the proposed method significantly outperforms other methods in key tasks such as question area segmentation mIoU of 0.94, handwriting formula recognition accuracy of 86.4%, and objective question grading F1 score of 97.5%. It also demonstrates stronger robustness in dealing with challenges of image quality, layout complexity, and writing standardization, with an average performance degradation rate of only 11.3%. This study confirms that the proposed deep visual understanding approach can effectively tackle the key challenges of automated handwritten homework processing and provides an efficient and reliable tool for educational informatization in terms of data collection and intelligent grading.

Cyber–physical systems (CPS) in healthcare represent a deep integration of computational intelligence, physical medical devices, and human-centric data, enabling continuous, adaptive, and personalized care. These systems combine real-time measurements, artificial intelligence (AI)-based analytics, and networked medical devices to monitor, predict, and optimize patient health outcomes. A key development in the field of CPS is the emergence of patient digital twins (DTs), virtual models of individual patients that simulate biological, behavioral, and social parameters. Using AI, DTs analyze complex medical and social data (genetics, lifestyle, environment, etc.) to support precise diagnosis and treatment planning. The implications of the bibliometric findings suggest that the field emerges from the conceptual phase, justifying the article’s emphasis on both the proposed architectures and their clinical validation. However, most research was conducted in computer science, engineering, and mathematics, rather than medicine and healthcare, suggesting an early stage of technological maturity. Leading countries were India, the United States, and China, but these countries did not have a high number of publications, nor did they record leading researchers or affiliations, suggesting significant research fragmentation. The most frequently observed Sustainable Development Goals indicate an industrial context. Reflecting insights from medical and social research, AI-based DT systems provide a holistic view of the patient, taking into account not only physiological states but also psychological and social well-being. These systems promote personalized therapy by dynamically adapting treatment based on real-time feedback from wearable sensors and electronic medical records. More broadly, CPS and DT systems increase healthcare system efficiency by reducing hospitalizations and supporting remote preventive care. Their implementation poses significant ethical and privacy challenges, particularly regarding data ownership, algorithm transparency, and patient autonomy.

The design practices on which the visual marketing management is founded have evolved into the data-driven and analytics-based systems of decisions. The following paper is an evaluation of the way AI will be used to radically transform the management of visual marketing by automated analysis, creating, optimization and controlling of visual content in online platforms. The research is premised on the visual communication and computational intelligence theory, and the authors present how computer vision, deep learning, and generative AI models can assist visual marketing to generate and implement strategic designs more effectively. It implies a fully experimental design that involves enormous image and video files created as a result of branding and advertising campaigns and social media promotion. It is measured in multi-level measures which include visual effectiveness, brand consistency, audience engagement and ROI. Empirical results of the research demonstrates that AI based visual marketing systems are much more useful in increasing relevance of content, emotion and cross platform compatibility of brands compared with manual systems or rule based systems. The outcomes also indicate measurable increase in efficacy of the campaigns, quicker design procedures, and better consistency of the enforcement of the brand identity.

Recent conservation of old sculptures is still a primary reinforcement to cultural heritage conservation; it has traditionally been based on the hand process that is subjective, irreversible and time consuming. The paper introduces a framework of AI-based restoration which incorporates multimodal data gathering, hybrid neural model, and expert-guided verification to attain accurate and ethically controlled digital restoration. The system makes use of LiDAR, CT, photogrammetry as well as multispectral imaging to acquire geometric and material information, which is processed with the help of a hybrid CNN-GAN-Transformer pipeline. The CNN derives structural, textual features, the GAN recreates the geometry that is missing and the Transformer imposes stylistic consistency with the help of knowledge-driven cultural embeddings. The quantitative analyses of three case studies of Roman marble, Chinese terracotta and Indian sandstone sculptures show that the framework is robust with 2530% reduction in Chamfer and Hausdorff distances, mean SSIM = 0.94, and cultural authenticity of above 4.3/5 by panels. Qualitative tests also prove that the restored outputs are both geometrical and culturally faithful. The architectural design enables the implementation of interactive, reversible, and transparent restoration processes to support the implementation of large-scale deployment of the modular architecture in museums, digital repositories, and AR/VR heritage platforms. In addition to performance, the framework focuses on ethical design of AI based on the concepts of human-in-the-loop testing, diversification of dataset, and documentation with provenance in consideration. Findings confirm the importance of AI as a cooperative stakeholder in the preservation of sculptural heritage of humankind, as an integration of computational intelligence and cultural accountability.

AI-assisted sculpture design is a radical melding of the conventional art craftsmanship and the innovative computational intelligence. The proposed study examines a hybrid form of creative ecosystem where sculptors work together with generative models, including GANs, diffusion systems, and mesh-generating neural networks to create sculptural, conceptually rich, structurally optimized, and culturally-infused sculptural entities. The suggested model will be based on the multimodal inputs which are hand-drawn sketches, 3D scans, material textures, and regional motifs which can allow the AI to be not only viewed as an automated tool but also as a collaborative contributor. Based upon the theories of human-machine collaboration and aesthetic cognition, the work presents how the concept of hybrid authorship redefines artistic intention, increases the speed of ideation, and facilitates experimentation with volumetric geometries that cannot be achieved in a field of manual work. The form of methodology is placed on the strict processing and annotation of sculptural data, native to curvature data, and surface anomalies, stylistic representation, and cultural emblem correlation. Moreover, a simulation layer of material consciousness is applied that assesses the reactions of stone, metals, clay, and composite, forecasts stressful regions, texture results, and manufacturability. The experiments show that there is a higher efficiency in design iteration, accuracy in integrating cultural motifs and physical plausibility of generated forms.

Intelligent movement tracking performing arts has become a paradigm shift in research, where computer intelligence is applied to creative performance. Conventional methods of motion analysis, most of which rely on manual observation, marker tracking systems, or single sensory modes, are incapable of tracking the subtleties, fluidity, and stylistic diversity of dances, theatre and performance. The recent innovations in artificial intelligence, multimodal sensing, and real-time analytics provide new opportunities to measure expressive movement with an unprecedented accuracy. This paper suggests an all-encompassing design that is based on optical cameras, inertial measurement units, depth sensors, and wearable devices and combines them with cutting-edge machine learning algorithms, including CNN-based pose estimators, graph convolution networks, and transformers. The system architecture has the focus of multimodal fusion, through which it is possible to consider the concurrent perception of visual, inertial, acoustic, and biomechanical signals to gain deeper insights into human movement. The processes of live performance environments, strong annotation of expressive and stylistic features, and deep learning architecture design based on the dynamics of performing arts are developed as a methodological pipeline. They have been applied to choreography analysis, automated assessment of movement-quality, intelligent systems of teaching dance and acting, and performance optimization using biomechanical feedback.

Neural Style Transfer (NST) has become a disruptive artistic process bridging the gap between computational intelligence and artistic expression, allowing the combination of content structures with styles inspired by a wide range of visual art pieces. The given research examines NST not as a technical algorithm, but as a modern aesthetic practice that widens the scope of digital art-making. The paper initially reviews the basic and advanced methods in artistic style transfer, which include algorithmic differences like Gram-matrix-based models, adaptive instance normalization, transformer based stylization and fast feed forward structures. It also compares these approaches and compares them with traditional fine-art methods to put the re-definitions of authorship, originality and artistic work into context. It uses a systematic approach to curating datasets, the choice of selection criteria of artistic exemplars and the design of neural architectures that trade-off style richness and content fidelity. In TensorFlow and PyTorch, the analysis of several style content trade-offs is performed focusing on the role of parameter optimization, selection of layers, and style-weight scaling in influencing the quality of expressions generated. The visual outcomes reveal how NST makes it possible to reinterpret artworks with delicate nuances of forms, textures, and coloration to create the artworks which are semantically consistent but stylistically abstract. The paper ends by critically analyzing limitations of NST, which can be summarized as, resolving of stylization, high computational cost, and inability to implement in real-time or generalized stylization in various artistic fields.

Background: Recent advancements in blood transfusion and transfusion medicine have increasingly integrated innovative technologies, including artificial intelligence (AI), machine learning, and computational intelligence. Despite numerous reviews on these topics, a comprehensive synthesis of the existing evidence is lacking. Objective: This narrative review of reviews aims to summarize and critically appraise the current literature on AI-driven and emerging technological approaches in blood transfusion, providing a structured overview for researchers and clinicians. Methods: A total of 19 reviews were selected through a systematic search strategy. Studies were assessed for methodological quality, scope, and clinical relevance, using adapted criteria from narrative review checklists. Data were extracted regarding the type of technology, application in transfusion medicine, study population, and reported outcomes. Results: The included reviews highlight several key domains: AI-assisted prediction of transfusion requirements, automated blood typing and crossmatching, advanced monitoring of blood products, and integration of computational models in blood banking workflows. Most studies reported promising applications but revealed substantial heterogeneity in methods, limited clinical validation, and variable reporting quality. Conclusions: AI and emerging technologies offer significant potential to improve the safety, efficiency, and personalization of blood transfusion. However, standardization of study designs, comprehensive validation, and robust reporting are essential to translate these innovations into routine clinical practice. This review of reviews provides a structured synthesis to guide future research and implementation strategies in transfusion medicine.

Artificial intelligence has proven to be a paradigm shift in preservation of cultural heritage and has made it possible to digitize vast portions of heritage, classify intelligently and semantically interconnect regional art forms. The paper introduces a full-fledged AI-based regional art mapping and preservation framework that incorporates various forms of multimodal data visual, textual, and geospatial data in an integrated cultural knowledge framework. The architecture has five layers, including data acquisition, AI analytics, knowledge graph integration, visualization interfaces and ethical governance. The implementation revolves around the Regional Art Knowledge Graph (RAKG) and Cultural Atlas Interface that establishes a semantic and interactive environment of exploring artistic relationships in terms of regions, styles, and time. Technical accuracy (A 1 = 92.3, F1 = 0.91, SSI = 0.84) and cultural authenticity (CAS = 8.7/10, RDI = 0.82) are good. Transparency and contextual fidelity are guaranteed by the presence of explainable AI systems and community engagement systems. The fusion of computational intelligence and human creativity allows the presented system to transform heritage preservation to a dynamic process that is participatory and ties the local traditions to a global digital future.

The international art market lends complex dynamics that interact with aesthetic perception, the cycle of economic activities and the mood of the investor, making it a difficult task to forecast prices. The paper presents a deep learning architecture that combines visual, contextual, and temporal data to predict the valuations of artworks with further accuracy. The study proposes a data engineering pipeline that is multimodal that includes curated image collections and structured information, including artist background, sales history, medium, and dimensions. A convolutional neural network (CNN) is used to produce high-level structure of artistic style and quality, whereas transformer and Long Short-Memory (LSTM) structures discover the temporal dynamics of price tendencies in the past. These modalities are amalgamated into a single embedding that is a combination of both visual and economic cues a fusion layer. This model is hyperparameter tuned and transferred learned with the help of pretrained encoders to optimize prediction accuracy and prevent overfitting with regularization measures. Findings indicate better performance compared to traditional econometric and regression models and better correlation with the real market trends and overall generalization across genres and time series. In addition to predictive ability, the framework offers interpretable information suggesting the impact of artistic qualities on valuation trends, therefore, linking computational intelligence to the art economics. The suggested system provides possible solutions in market analytics, auction prediction, and digital art investment platforms, which add to the development of data-driven decision-making in the creative economy.

Context. To enhance the performance of numerical optimization techniques, hybrid approaches integrating probabilistic modeling algorithms with annealing simulation have been introduced. These include Bayesian optimization, Markov-based strategies, and extended compact genetic algorithms, each augmented by annealing mechanisms. Such methods enable more precise search trajectories without requiring fitness function transformation, owing to their ability to explore the global search space in early iterations and refine the directionality of search in later stages.Objective. The research aims to improve the effectiveness of parameter identification within approximation models of financial indicators by applying metaheuristic algorithms that incorporate probabilistic modeling and annealing-based simulation in intelligent computing systems.Method. This study employs metaheuristic techniques grounded in probabilistic modeling and annealing-based simulation to enhance the accuracy and efficiency of parameter estimation within economic indicator approximation frameworks. Specifically, it introduces three hybrid strategies: Bayesian-based optimization integrated with annealing simulation, Markov-driven optimization enhanced by annealing, and an extended compact genetic algorithm coupled with annealing mechanisms. These methods enhance the accuracy of the search process by exploring the entire search space in initial iterations and refining the search direction in final iterations. The Bayesian optimization method employs a Bayesian network for structured search and solution refinement. The Markov optimization method integrates Gibbs quantization within a Markov network to improve search precision. The extended compact genetic algorithm utilizes limit distribution models to generate optimal solutions. These methods eliminate the need for fitness function transformation, optimizing computational efficiency. The proposed techniques expand the application of metaheuristics in intelligent economic computer systems.Results. The implemented optimization strategies significantly enhanced the precision of parameter estimation within intelligent financial computing frameworks. The combination of probabilistic models and annealing simulation enhanced search efficiency without requiring fitness function transformation.Conclusions. The proposed method expands the application of metaheuristics in economic modeling, increasing computational effectiveness. Further research should explore their implementation across diverse artificial intelligence problems.

The development of new technologies, especially artificial intelligence, creates enormous opportunities for humanity on the way to further progress and development. Nowadays, no one is surprised by the fact that robots work much more efficiently than humans, and the resources of the Internet are many times greater than human memory. Artificial intelligence will also inevitably enter the everyday life of the modern world, and there is no need to fear that the development of computer intelligence will soon surpass the potential of the human mind. However, already at this stage it is necessary to distinguish digital international law by outlining the ethical and legal framework for the operation of artificial intelligence systems. Without a doubt, such a framework must take into account human rights, data protection, the rule of law, protection of intellectual property, and principles of liability for damage caused by the actions of computer intelligence. The current European achievements (within the Council of Europe and the EU) in the field of regulation of initial standards for the principles of functioning of systems based on artificial intelligence are pioneering on a global scale and for this reason it is worth making a synthetic analysis of them. This is also the aim of this study.

Computational Intelligence for Predicting the Strength of High-Performance Concrete: Ensemble and Non-ensemble Machine Learning Framework