Search Results (813)

Financial time series prediction remains a challenging task due to the inherent non-stationarity, noise, and complex temporal dependencies present in market data. Traditional forecasting methods often fail to capture the multifaceted nature of financial markets, where temporal proximity, trend dynamics, and volatility patterns simultaneously influence price movements. To address these limitations, this paper proposes the Enhanced Multi-Aspect Transformer (EMAT), a novel deep learning architecture specifically designed for stock market prediction. EMAT incorporates a Multi-Aspect Attention Mechanism that simultaneously captures temporal decay patterns, trend dynamics, and volatility regimes through specialized attention components. The model employs an encoder–decoder architecture with enhanced feed-forward networks utilizing SwiGLU activation, enabling superior modeling of complex non-linear relationships. Furthermore, we introduce a comprehensive multi-objective loss function that balances point-wise prediction accuracy with volatility consistency. Extensive experiments on multiple stock market datasets demonstrate that EMAT consistently outperforms a wide range of state-of-the-art baseline models, including various recurrent, hybrid, and Transformer architectures. Our ablation studies further validate the design, confirming that each component of the Multi-Aspect Attention Mechanism makes a critical and quantifiable contribution to the model’s predictive power. The proposed architecture’s ability to simultaneously model these distinct financial characteristics makes it a particularly effective and robust tool for financial forecasting, offering significant improvements in accuracy compared to existing approaches. Full article

The global transition toward renewable energy sources has intensified in response to escalating environmental challenges. Nevertheless, the inherent intermittency and instability of renewable energy necessitate the development of reliable energy storage technologies. Supercapacitors are particularly notable for their high specific capacitance, rapid charge and discharge capability, and exceptional cycling stability. Concurrently, the increasing demand for efficient and sustainable energy storage systems has stimulated interest in multifunctional electrode materials that integrate electrocatalytic activity with electrochemical energy storage. Two-dimensional transition metal dichalcogenides (TMDs), owing to their distinctive layered structures, large surface areas, phase state, energy band structure, and intrinsic electrocatalytic properties, have emerged as promising candidates to achieve dual functionality in electrocatalysis and electrochemical energy storage for asymmetric supercapacitors (ASCs). Specifically, their unique electronic properties and catalytic characteristics promote reversible Faradaic reactions and accelerate charge transfer kinetics, thus markedly enhancing charge storage efficiency and energy density. This review highlights recent advances in TMD-based multifunctional electrodes. It elucidates mechanistic correlations between intrinsic electronic properties and electrocatalytic reactions that influence charge storage processes, guiding the rational design of high-performance ASC systems. Full article

The medical application of human embryonic stem cell technology has sparked ethical controversies, with the core issue being whether human embryos possess the same right to life as humans. According to classical Confucianism, humans are born from the essential Qi 精氣 (jingqi) of heaven and earth, making them the noblest beings in the world. Human embryos are the simple form of human life in its early stages, and as living human beings, they should therefore possess the legitimacy and justification to life. Confucianism advocates benevolence toward people 仁民 (renmin) and love for things 愛物 (aiwu) distinguishing between benevolence and love: benevolence toward people is benevolence, while love for all things is love. How people treat one another is how they should treat human embryos. Things exist to serve humanity; humans may utilise things but must not be treated as tools. Embryo life must not be harmed or sacrificed for the sake of saving human life. One should show benevolence to people and love to things. Therefore, the attitude toward human embryos should be “benevolence.” Human embryos inherently possess the potential to become human beings and do not require medical intervention to demonstrate their value. However, when humans extract and utilise stem cells from human embryos for their own benefit, this is tantamount to treating the embryos as things and reducing them to the status of things, thereby blurring the ethical boundaries between humans and things and disrupting the distinction between the recipients of benevolence and love. The extraction of human embryonic stem cells is ultimately an artificial technological achievement. Humans are not superior beings to heaven, and such practices must be confined within the moral framework of technological ethics and bioethics. Notwithstanding the technological advancements that have furnished humans with contemporary instruments, the necessity for a sense of awe for the heaven remains. Full article

Global Navigation Satellite System (GNSS) spoofing, which manipulates PVT solutions through false measurements, increasingly threatens GNSS reliability and user safety. However, most existing simulator-based spoofers, constrained by their inability to access real-time ionospheric information (e.g., Global Ionosphere Maps, GIMs) from external sources, struggle to replicate authentic total electron content (TEC) along each signal propagation path accurately and in a timely manner. In contrast, widespread dual-frequency (DF) receivers with access to the internet can validate local TEC measurements against external references, establishing a pivotal spoofing detection distinction. Here, we propose an Ionospheric Information-Assisted Spoofing Detection Technique (IIA-SDT), exploiting the inherent consistency between TEC values derived from DF pseudo-range measurements and external references in spoofing-free scenarios. Spoofing probably disrupts this consistency: in simulator-based full-channel spoofing where all channels are spoofed, the inaccuracies of the offline ionospheric model used by the spoofer inevitably cause TEC mismatches; in partial-channel spoofing where the spoofer fails to control all channels, an unintended PVT deviation is induced, which also causes TEC deviations due to the spatial variation of the ionosphere. Basic principles and theoretical analysis of the proposed IIA-SDT are elaborated in the paper. Simulations using ionospheric data collected from 2023 to 2024 at a typical mid-latitude location are conducted to evaluate IIA-SDT performance under various parameter configurations. With a window length of 5 s and satellite number of 8, the annual average detection probability approximates 75% at a false alarm rate of $1\times {10}^{-3}$, with observable temporal variations. Field experiments across multiple scenarios further validate the spoofing detection capability of the proposed method. Full article

This paper presents the design, manufacturing, testing, and validation of the MASKARA (Multiple Apertures for high-resolution SAR based on Ka-band Reflectarray) Breadboard Model (BBM), a large Ka-band reflectarray antenna developed for Synthetic Aperture Radar (SAR) applications. The BBM features a dual-offset antenna configuration intended for a high-resolution, wide-swath SAR instrument. At the core of the system is a 1.5 m × 0.55 m reflectarray operating between 35.5–36.0 GHz in the Ka-band. To our knowledge, this is the first demonstration of a reflectarray antenna designed to support two distinct modes of operation, exploiting the inherent advantages of reflectarrays—such as reduced cost and compact stowage—over traditional solutions. The antenna provides a high-resolution mode requiring a higher-gain beam in one polarization and a low-resolution mode covering a larger swath with broader beam coverage in the orthogonal polarization. The design process follows a holistic, multidisciplinary approach, integrating RF and thermomechanical considerations through iterative and concurrent design reviews. The BBM has been successfully manufactured and experimentally tested, and the measurement results show good agreement with simulations, confirming the validity of the proposed concept and demonstrating its potential for future high-performance SAR missions. Full article

Traditional anti-seismic methods are constrained by high construction costs and the potential for severe structural damage under earthquakes. Energy dissipation technology provides an effective solution for structural earthquake resistance by incorporating energy-dissipating devices within structures to actively absorb seismic energy. However, existing research lacks in-depth analysis of the influence of energy dissipation devices’ placement on structural dynamic response. Therefore, this study investigates the seismic mitigation effectiveness of viscous dampers in multi-storey frame structures and their optimal placement strategies. A comprehensive parametric investigation was conducted using a representative three-storey steel-frame kindergarten facility in Shandong Province as the prototype structure. Advanced finite element modeling was implemented through ETABS software to establish a high-fidelity structural analysis framework. Based on the supplemental virtual damping ratio seismic design method, damping schemes were designed, and the influence patterns of different viscous damper arrangement schemes on the seismic mitigation effectiveness of multi-storey frame structures were systematically investigated. Through rigorous comparative assessment of dynamic response characteristics and energy dissipation mechanisms inherent to three distinct energy dissipation device deployment strategies (perimeter distribution, central concentration, and upper-storey localization), this investigation delineates the governing principles underlying spatial positioning effects on structural seismic mitigation performance. This comprehensive investigation elucidates several pivotal findings: damping schemes developed through the supplemental virtual damping ratio-based design methodology demonstrate excellent applicability and predictive accuracy. All three spatial configurations effectively attenuate structural seismic response, achieving storey shear reductions of 15–30% and inter-storey drift reductions of 19–28%. Damper spatial positioning critically influences mitigation performance, with perimeter distribution outperforming central concentration, while upper-storey localization exhibits optimal overall effectiveness. These findings validate the engineering viability and structural reliability of viscous dampers in multi-storey frame applications, establishing a robust scientific foundation for energy dissipation technology implementation in seismic design practice. Full article

The increasing presence of emerging contaminants (ECs) has attracted considerable attention due to their potential harm to human health and ecosystems. Graphitic carbon nitride (g-C₃N₄), a semiconductor devoid of metals, stands out due to its distinctive optical properties and strong resistance to chemical degradation, which holds significant promise in the photocatalytic degradation of ECs. However, the inherent limitations of g-C₃N₄, such as its reduced specific surface area and the swift recombination of photogenerated electron-hole pairs, have prompted extensive research on modification strategies to enhance its photocatalytic performance. Current research on g-C₃N₄-based materials is often constrained in scope, with most reviews focusing solely on modification strategies or its application in degrading a single category of emerging contaminants (ECs). In this review, a systematic overview of synthesis methods and advanced modification strategies for g-C₃N₄-based materials is discussed, highlighting their recent advances in the photocatalytic degradation of various ECs using g-C₃N₄-based materials, which underscores their potential for environmental remediation. Moreover, this article critically examines the current challenges and outlines future research directions, with particular emphasis on integrating artificial intelligence and machine learning to accelerate the development of g-C₃N₄-based photocatalysts and optimize degradation processes, thereby promoting their efficient application in the photocatalytic degradation of ECs. Full article

This study extends the Theory of Functional Connections, previously applied to constraints specified at discrete points, to encompass continuous integral constraints of the form ${\displaystyle {\int }_{x_0}^{x_f}f(x,t)\mathrm{d}x=\mathcal{I}\left(t\right)}$, where $\mathcal{I}\left(t\right)$ can be a constant, a prescribed function, or an unknown function to be estimated through optimization. The framework of continuous integral constraints is developed within the context of initial value problems (IVP) and boundary value problems (BVP). To demonstrate the effectiveness of this analytical approach, examples validate the method and highlight distinctions between satisfying continuous integral constraints via simple interpolation versus functional interpolation. A limitation of the proposed approach is the inability to inherently enforce inequality constraints, such as the positivity constraint $f(x,t)\ge 0$, for modeling probability density functions in classical mechanics. Despite this, numerical experiments on boundary-value problems rarely result in negative values, indicating that the issue occurs infrequently. However, a mitigation strategy based on non-negative least-squares methods combined with Bernstein polynomials is proposed to address these rare cases. This approach is validated through an additional numerical test, demonstrating its efficacy in ensuring nonnegativity when required. Full article

The widespread adoption of deep learning (DL) models in neuroscience research has introduced a fundamental epistemological paradox: while these models demonstrate remarkable performance in pattern recognition and prediction tasks, their inherent opacity contradicts neuroscience’s foundational goal of understanding biological mechanisms. This review article examines the growing trend of using DL models to interpret neural dynamics and extract insights about brain function, arguing that the black box nature of these models fundamentally undermines their utility for mechanistic understanding. We explore the distinction between computational performance and scientific explanation, analyze the limitations of current interpretability techniques, and discuss the implications for neuroscience research methodology. We propose that the field must critically evaluate whether DL models can genuinely contribute to our understanding of neural processes or whether they merely provide sophisticated curve-fitting tools that obscure rather than illuminate the underlying biology. Full article

The globalization of hardware design flows has increased the risk of Hardware Trojan (HT) insertion during the design phase. Graph-based learning methods have shown promise for HT detection at the Register Transfer Level (RTL). However, most existing approaches rely on representing RTL designs through a single graph structure. This single-view modeling paradigm inherently constrains the model’s ability to perceive complex behavioral patterns, consequently limiting detection performance. To address these limitations, we propose GNN-MFF, an innovative multi-view feature fusion model based on Graph Neural Networks (GNNs). Our approach centers on joint multi-view modeling of RTL designs to achieve a more comprehensive representation. Specifically, we construct complementary graph-structural views: the Abstract Syntax Tree (AST) capturing structure information, and the Data Flow Graph (DFG) modeling logical dependency relationships. For each graph structure, customized GNN architectures are designed to effectively extract its features. Furthermore, we develop a feature fusion framework that leverages a multi-head attention mechanism to deeply explore and integrate heterogeneous features from distinct views, thereby enhancing the model’s capacity to structurally perceive anomalous logic patterns. Evaluated on an extended Trust-Hub-based HT benchmark dataset, our model achieves an average F1-score of 97.08% in automated detection of unseen HTs, surpassing current state-of-the-art methods. Full article

Addressing the critical challenge in the differential diagnosis of severe inflammatory lung diseases, we propose a novel methodology for the analysis of macrophage surface receptors, CD68 and CD206, using specific non-antibody ligands. We developed a non-antibody alternative for the fluorometric detection of CD68+ cells, focusing on macrophages as key functional markers in inflammatory processes. Our marker based on dioleylphosphatidylserine (DOPS), a specific ligand to CD68, was incorporated into a liposomal delivery system. The specificity of this DOPS-based ligand can be precisely modulated by the liposome’s composition and the polyvalent presentation of the ligand. We synthesized a series of fluorescently-labeled DOPS-based ligands and developed a liposome-based sandwich fluorometric assay. This assay enables the isolation and quantification of CD68 receptor presence from bronchoalveolar lavage fluid (BALF). The results confirmed the specific binding of DOPS/lecithin liposomes to CD68+ cells compared to control lecithin systems. Furthermore, the incorporation of PEGylated ‘stealth’ liposomes significantly enhanced binding specificity and facilitated the generation of distinct binding profiles, which proved valuable in differentiating various inflammatory conditions. This approach yielded unique binding profiles of PS-based ligands to CD68+ cells, which varied significantly among a broad range of respiratory conditions, including primary ciliary dyskinesia, bronchial asthma, bronchitis, bacterial infection, pneumonia, and bronchiectasis. Confocal Laser Scanning Microscopy demonstrated selective binding and intracellular localization of the DOPS-based marker within CD68+ macrophages from BALF samples of patients with bronchitis or asthma. The binding parameters of this multivalent composite ligand with the CD68 receptor are comparable to those of antibodies. The inherent binding specificity of phosphatidylserine may offer a sufficient and viable alternative to conventional antibodies. Our results demonstrate the remarkable potential of this novel DOPS-based assay as a complementary tool for the developing non-antibody-based systems for the differential diagnosis of the respiratory diseases, warranting further investigation in larger clinical studies. Full article

The increase in Internet of Things (IoT) devices has introduced significant security challenges, primarily due to their inherent constraints in computational power, memory, and energy. This study provides a comparative performance analysis of selected modern cryptographic algorithms on a resource-constrained IoT platform, the Nordic Thingy:53. We evaluated a set of ciphers including the NIST lightweight standard ASCON, eSTREAM finalists Salsa20, Rabbit, Sosemanuk, HC-256, and the extended-nonce variant XChaCha20. Using a dual test-bench methodology, we measured energy consumption and performance under two distinct scenarios: a low-data-rate Bluetooth mesh network and a high-throughput bulk data transfer. The results reveal significant performance variations among the algorithms. In high-throughput tests, ciphers like XChaCha20, Salsa20, and ASCON32 demonstrated superior speed, while HC-256 proved impractically slow for large payloads. The Bluetooth mesh experiments quantified the direct relationship between network activity and power draw, underscoring the critical impact of cryptographic choice on battery life. These findings offer an empirical basis for selecting appropriate cryptographic solutions that balance security, energy efficiency, and performance requirements for real-world IoT applications. Full article

The prediction of stock price trends is of vital importance for maintaining the stability of the financial market, optimizing resource allocation and preventing systemic risks. To ensure the practical application value of the prediction model, it is necessary to maintain prediction accuracy while ensuring that the output results of the model are interpretable, enabling decision-makers to understand and verify the prediction basis. Belief Rule Base (BRB) models, grounded in IF-THEN rule semantics, offer inherent interpretability. However, optimizing BRB models can erode this interpretability, and they are susceptible to combinatorial explosion in multi-attribute scenarios, disrupting the structural symmetry and escalating model complexity. To address these challenges while preserving both accuracy and interpretability symmetry, this paper proposes a new method based on hierarchical Belief Rule Base with balanced accuracy and interpretability (HBRB-b) for stock price trend prediction. First, a hierarchical model structure is constructed to overcome the rule combinatorial explosion problem, ensuring initial structural symmetry and interpretability. Second, several interpretability criteria specifically designed for stock prediction and compatible with maintaining model balance during optimization are proposed to guide the modeling process. Finally, an improved Whale Optimization Algorithm is proposed, incorporating constraints to preserve the interpretability symmetry throughout the optimization process. A case study validates the model’s effectiveness in stock price trend prediction. Comparative results demonstrate that the HBRB-b-based model achieves a favorable symmetry between prediction accuracy and model interpretability, offering distinct advantages in both aspects. Full article

This essay explores whether romantic exclusivity is more than a cultural choice, suggesting it might be built into the very structure of love. Turning away from typical sociological or psychological explanations, I place classical philosophy in direct conversation with contemporary thinkers like Natasha McKeever, Christopher Bennett, and Carrie Jenkins to investigate this question. I argue that a synthesis of three distinct philosophical frameworks reveals exclusivity as a structural requirement for romantic love in its deepest sense. First, drawing on Kant, I suggest that love’s demand for a totalizing cognitive synthesis of two lives runs into a transcendental barrier when attempted with more than one person. Second, I use Scheler’s phenomenology to argue that the deep, sustained attention required for love’s unique power of value revelation is inherently diluted across multiple partners. Third, I introduce Schopenhauer’s metaphysics to posit that divided romantic striving contradicts the indivisible nature of the Will. I also briefly touch on how thinkers like Kierkegaard and Levinas reinforce this theme of existential singularity. Taken together, this synthesis does not condemn non-monogamous relationships but reframes the debate. It suggests that what we call “romantic love” may be structurally distinct from other valuable forms of intimacy. The powerful pull toward exclusivity, then, might not be a mere social script but may reflect the fundamental architecture of consciousness, valuation, and being itself. Full article

Traditional spam detection methodologies often neglect user privacy preservation, potentially incurring data leakage risks. Furthermore, current federated learning models for spam detection face several critical challenges: (1) data heterogeneity and instability during server-side parameter aggregation, (2) training instability in single neural network architectures leading to mode collapse, and (3) constrained expressive capability in multi-module frameworks due to excessive complexity. These issues represent fundamental research pain points in federated learning-based spam detection systems. To address this technical challenge, this study innovatively integrates federated learning frameworks with multi-feature fusion techniques to propose a novel spam detection model, FPW-BC. The FPW-BC model addresses data distribution imbalance through the FedProx aggregation algorithm and enhances stability during server-side parameter aggregation via a horse-racing selection strategy. The model effectively mitigates limitations inherent in both single and multi-module architectures through hierarchical multi-feature fusion. To validate FPW-BC’s performance, comprehensive experiments were conducted on six benchmark datasets with distinct distribution characteristics: CEAS, Enron, Ling, Phishing_email, Spam_email, and Fake_phishing, with comparative analysis against multiple baseline methods. Experimental results demonstrate that FPW-BC achieves exceptional generalization capability for various spam patterns while maintaining user privacy preservation. The model attained 99.40% accuracy on CEAS and 99.78% on Fake_phishing, representing significant dual improvements in both privacy protection and detection efficiency. Full article