Search Results (1,811)

In the big data era, understanding and influencing consumer behavior in digital marketing increasingly relies on large-scale data and AI-driven analytics. This narrative, concept-driven review examines how big data technologies and machine learning reshape consumer behavior analysis across key decision-making areas. After outlining the theoretical foundations of consumer behavior in digital settings and the main data and AI capabilities available to marketers, this paper discusses five application domains: personalized marketing and recommender systems, dynamic pricing, customer relationship management, data-driven product development and fraud detection. For each domain, it highlights how algorithmic models affect targeting, prediction, consumer experience and perceived fairness. This review then turns to synthetic data as a privacy-oriented way to support model development, experimentation and scenario analysis, and to dark data as a largely underused source of behavioral insight in the form of logs, service interactions and other unstructured records. A discussion section integrates these strands, outlines implications for digital marketing practice and identifies research needs related to validation, governance and consumer trust. Finally, this paper sketches future directions, including deeper integration of AI in real-time decision systems, increased use of edge computing, stronger consumer participation in data use, clearer ethical frameworks and exploratory work on quantum methods. Full article

Anomalous transaction behaviors in cryptocurrency markets exhibit high concealment, substantial diversity, and strong cross-modal coupling, making traditional rule-based or single-feature analytical methods insufficient for reliable detection in real-world environments. To address the research focus, a data-centric multimodal anomaly detection framework integrating generative augmentation, latent distribution modeling, and dual-branch real-time detection is proposed. The method employs a generative adversarial network with feature-consistency constraints to mitigate the scarcity of fraudulent samples, and adopts a multi-domain variational modeling strategy to learn the latent distribution of normal behaviors, enabling stable anomaly scoring. By combining the long-range temporal modeling capability of Transformer architectures with the sensitivity of online clustering to local structural deviations, the system dynamically integrates global and local information through an adaptive risk fusion mechanism, thereby enhancing robustness and real-time detection capability. Experimental results demonstrate that the generative augmentation module yields substantial improvements, increasing the recall from $0.421$ to $0.671$ and the F1-score to $0.692$. In anomaly distribution modeling, the multi-domain VAE achieves an area under the curve (AUC) of $0.854$ and an F1-score of $0.660$, significantly outperforming traditional One-Class SVM and autoencoder baselines. Multimodal fusion experiments further verify the complementarity of the dual-branch detection structure, with the adaptive fusion model achieving an AUC of $0.884$, an F1-score of $0.713$, and reducing the false positive rate to $0.087$. Ablation studies show that the complete model surpasses any individual module in terms of precision, recall, and F1-score, confirming the synergistic benefits of its integrated components. Overall, the proposed framework achieves high accuracy and high recall in data-scarce, structurally complex, and latency-sensitive cryptocurrency scenarios, providing a scalable and efficient solution for deploying data-centric artificial intelligence in financial security applications. Full article

An Integrated Energy System (IES) integrates electricity, heat, and natural gas, optimizing energy use and management efficiency. These systems connect to a Virtual Power Plant (VPP) for demand response dispatch in the electricity market. However, the impact of VPP load on the IES is often overlooked, which can limit the IES’s effective market participation and stability. To address this issue, this study introduces a two-layer collaborative model to coordinate VPP scheduling for multiple IES units, aiming to improve collaboration efficiency. The upper level involves the VPP setting electricity prices based on load conditions, guiding IES units to adjust their market strategies. At the lower level, the model encourages integration and optimization of different energy types within the IES through enhanced energy interactions. Additionally, the application of the Shapley value method ensures fair benefit distribution among all IES members. This approach supports equitable economic outcomes for all participants in the energy market. The model employs a multi-strategy improved Dung Beetle Optimizer (FSGDBO) combined with commercial solver techniques for efficient problem-solving. Experimental results demonstrate that the model significantly enhances the VPP’s peak-shaving and valley-filling capabilities while preserving the economic interests of the IES alliances, thereby boosting overall energy management effectiveness. Full article

This study employs global optimization techniques to examine optimal vaccine pricing strategies that consider quantity discounts and vaccine distribution equity, under centralized procurement by group purchasing organizations. Based on the economic characteristics of the vaccine market, a mathematical programming model incorporates the payment capacities and willingness to pay of different member countries, minimizing the maximum adjusted price disparities across pricing tiers and thereby enhancing the overall fairness of vaccine distribution. To further reduce computational complexity and enhance practical applicability, this study improves the model by reducing the number of binary variables. Experimental analysis is conducted using real-world data from the Vaccine Alliance (Gavi) and the Pan American Health Organization (PAHO). The results show that the improved model reduces computation time by over 30% on average and demonstrates effective control over price differentiation across various pricing tiers and parameter settings. Full article

Defect detection on Printed Circuit Boards (PCBs) constitutes a pivotal component of the quality control system in electronics manufacturing. However, owing to the intricate circuitry structures on PCB surfaces and the characteristics of defects—specifically their minute scale, irregular morphology, and susceptibility to background texture interference—existing generic deep learning models frequently fail to achieve an optimal equilibrium between detection accuracy and inference speed. To address these challenges, this study proposes MDEB-YOLO, a lightweight real-time detection network tailored for PCB micro-defects. First, to enhance the model’s perceptual capability regarding subtle geometric variations along conductive line edges, we designed the Efficient Multi-scale Deformable Attention (EMDA) module within the backbone network. By integrating parallel cross-spatial channel learning with deformable offset networks, this module achieves adaptive extraction of irregular concave–convex defect features while effectively suppressing background noise. Second, to mitigate feature loss of micro-defects during multi-scale transformations, a Bidirectional Residual Multi-scale Feature Pyramid Network (BRM-FPN) is proposed. Utilizing bidirectional weighted paths and residual attention mechanisms, this network facilitates the efficient fusion of multi-view features, significantly enhancing the representation of small targets. Finally, the detection head is reconstructed based on grouped convolution strategies to design the Lightweight Grouped Convolution Head (LGC-Head), which substantially reduces parameter volume and computational complexity while maintaining feature discriminability. The validation results on the PKU-Market-PCB dataset demonstrate that MDEB-YOLO achieves a mean Average Precision (mAP) of 95.9%, an inference speed of 80.6 FPS, and a parameter count of merely 7.11 M. Compared to baseline models, the mAP is improved by 1.5%, while inference speed and parameter efficiency are optimized by 26.5% and 24.5%, respectively; notably, detection accuracy for challenging mouse bite and spur defects increased by 3.7% and 4.0%, respectively. The experimental results confirm that the proposed method outperforms state-of-the-art approaches in both detection accuracy and real-time performance, possessing significant value for industrial applications. Full article

The development and application of therapeutic oligonucleotides, such as siRNA, miRNA, ASOs and aptamers, is a rapidly growing field in biomedicine. These molecules are undergoing extensive preclinical and clinical testing, and the market for synthetic RNA drugs is expanding. However, several challenges remain, including targeted delivery and high costs associated with development, screening and production. One significant advance has been the creation of GalNAc-conjugates, which selectively target ASGPR and deliver oligonucleotides to hepatocytes. Although these conjugates have shown promising results, their widespread use is limited by the lack of effective synthesis methods. Thus, the development of new methods for the synthesis of ligand-oligonucleotide conjugates is an important task to which this study is devoted. In this study, we created a library of siRNA conjugates with the GalNAc L-96 ligand to suppress the expression of the PCSK9 gene associated with elevated LDL and an increased risk of developing cardiovascular diseases. The selection of the most effective siRNA molecules was carried out using an algorithm previously developed by our research group, which considers thermodynamic stability, predicted specificity and effectiveness. To experimentally confirm the effectiveness of conjugates, an in vitro model based on the cultivation of hepatocyte cells was developed. Optimization of the conjugate synthesis process has significantly reduced the cost of manufacturing technology, which creates the potential for efficient scaling of synthesis for transfer and application in the pharmaceutical industry. The results of the study showed that the development of the siRNA sequence optimized in silico resulted in a significant increase in the inhibitory effect of the GalNAc-siRNA conjugate compared to a compound similar to a commercial drug. Full article

This study aims to optimize the voltage profile of the grid by obtaining an optimum level of reactive power support from photovoltaic (PV) plants, thereby enhancing the efficiency of PV systems in power distribution networks and ensuring grid stability. Initially, voltage profiles in the sector, together with the structure and operating principles of PV plants, were considered in detail. Subsequently, the limits of reactive power support that can be provided by PV plants were determined. Then, the optimum levels of reactive power from the plants were determined using particle swarm optimization, genetic algorithm, Jaya algorithm, and firefly algorithm separately. The algorithms were tested through simulations conducted on a power distribution system operator in Türkiye. Additionally, a Modbus-based communication application was developed and tested, as a feasibility demonstration, to verify PV inverter accessibility and the capability of remotely writing reactive power reference setpoints. The quantitative optimization results reported in this manuscript are obtained from DIgSILENT PowerFactory simulations using the actual feeder model and time-series profiles. The results have revealed that PV plants can be effectively utilized as reactive power compensators to contribute to the operation of the grid under more ideal voltage profile conditions. In Türkiye, there is no regulatory or market mechanism to support reactive power provision from PV plants. Therefore, this study is novel in the Turkish market. The experimental results confirm that power generation from renewable energy can provide reactive support effectively when needed, which reveals that this approach is both technically feasible and practically relevant. Full article

To meet the development needs of high-rise timber structures, current cross-laminated timber (CLT) shear walls typically feature a single-layer thickness of 35 mm with more than three laminations in the stack. However, such thickness easily leads to resource waste in small-scale residential buildings, while increasing transportation and hoisting costs, which is not conducive to the prefabrication and lightweight development of timber structures. To adapt to the development trend of China’s timber structure market towards public buildings such as cultural and tourism projects and small-scale residential buildings including new rural housing renovation, this study focuses on thin CLT shear walls with an overall thickness of 48 mm (16 mm per layer) and conducts research on their lateral load-bearing performance. Monotonic lateral static load tests and finite element (FE) simulations were carried out on thin CLT shear walls without openings, with different opening areas, and with the same opening area but different positions. A corresponding FE model was established and validated, with a focus on analyzing the influence of opening parameters on the shear performance of the walls. The research results show that wall openings significantly reduce the bearing capacity and shear stiffness of the walls: compared with the wall without openings, the ultimate load and shear stiffness of the walls with openings decrease by 20.4–28.6% and 36.3–42.3%, respectively. Among them, increasing the opening height has a more obvious weakening effect on the bearing capacity; for the same opening area, a wider opening results in a more significant decrease in stiffness. The FE model exhibits reliable accuracy, with the error between the experimental and simulation results in the elastic stage controlled within 10%, and the influence of the under-wall support on the shear stiffness is relatively small. Opening parameters have a prominent impact on the stiffness of the wall in the elastic stage, and the influence of the opening position is more critical—the smaller the distance from the opening to the top of the wall, the more obvious the decrease in overall stiffness. Full article

Aim: Stock price prediction remains a highly challenging task due to the complex and nonlinear nature of financial time series data. While deep learning (DL) has shown promise in capturing these nonlinear patterns, its effectiveness is often hindered by the low signal-to-noise ratio inherent in market data. This study aims to enhance the stock predictive performance and trading outcomes by integrating Singular Spectrum Analysis (SSA) with deep learning models for stock price forecasting and strategy development on the Australian Securities Exchange (ASX)50 index. Method: The proposed framework begins by applying SSA to decompose raw stock price time series into interpretable components, effectively isolating meaningful trends and eliminating noise. The denoised sequences are then used to train a suite of deep learning architectures, including Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), and hybrid CNN-LSTM models. These models are evaluated based on their forecasting accuracy and the profitability of the trading strategies derived from their predictions. Results: Experimental results demonstrated that the SSA-DL framework significantly improved the prediction accuracy and trading performance compared to baseline DL models trained on raw data. The best-performing model, SSA-CNN-LSTM, achieved a Sharpe Ratio of 1.88 and a return on investment (ROI) of 67%, indicating robust risk-adjusted returns and effective exploitation of the underlying market conditions. Conclusions: The integration of Singular Spectrum Analysis with deep learning offers a powerful approach to stock price prediction in noisy financial environments. By denoising input data prior to model training, the SSA-DL framework enhanced signal clarity, improved forecast reliability, and enabled the construction of profitable trading strategies. These findings suggested a strong potential for SSA-based preprocessing in financial time series modeling. Full article

Mango is a high-value agricultural commodity, and accurate and efficient appearance quality grading and defect inspection are critical for export-oriented markets. This study proposes a dual-pipeline deep learning framework for automated mango assessment, in which surface defect classification and quality grading are jointly implemented within a unified inspection system. For defect assessment, the task is formulated as a multi-label classification problem involving five surface defect categories, eliminating the need for costly bounding box annotations required by conventional object detection models. To address the severe class imbalance commonly encountered in agricultural datasets, a copy–paste-based image synthesis strategy is employed to augment scarce defect samples. For quality grading, mangoes are categorized into three quality levels. Unlike conventional CNN-based approaches relying solely on spatial-domain information, the proposed framework integrates decision-level fusion of spatial-domain and frequency-domain representations to enhance grading stability. In addition, image preprocessing is investigated, showing that adaptive contrast enhancement effectively emphasizes surface textures critical for quality discrimination. Experimental evaluations demonstrate that the proposed framework achieves superior performance in both defect classification and quality grading compared with existing detection-based approaches. The proposed classification-oriented system provides an efficient and practical integrated solution for automated mango assessment. Full article

This research experimentally investigates the association between multidimensional risk factors and economic growth, quantified by GDP as a partial indicator of advancement towards economically relevant Sustainable Development Goals (SDGs). This research experimentally investigates the correlation between multidimensional risk variables and economic growth, quantified by GDP as a partial indicator of advancement towards economically relevant Sustainable Development Goals (SDGs) in Saudi Arabia, particularly in alignment with the objectives of Saudi Vision 2030. This study utilizes annual data from 1990 to 2024 and employs the Autoregressive Distributed Lag (ARDL) bounds testing approach to examine the short-run and long-run relationships between economic growth, as measured by GDP, and five key risk dimensions: governance effectiveness, financial development, environmental pressure, human capital, and oil price volatility, which act as proxies for risk dimensions. The main contribution of this study is the integration of these governance, financial, environmental, human capital, and oil price risk factors into a single ARDL framework for Saudi Arabia from 1990 to 2024, using GDP growth as a proxy for progress toward SDGs within the Saudi Vision 2030 context, addressing gaps in prior studies that focus on individual determinants. The empirical evidence indicates a long-term cointegration relationship among the variables. Our findings indicate that government effectiveness and investment in human capital are important positive factors associated with long-term economic growth, thereby validating the importance of institutional improvements and educational expenditures. In contrast, fluctuations in oil prices and environmental pressures are linked to adverse association, highlighting issues related to resource dependency and ecological degradation. Financial development exhibits a negative long-run association, indicating potential inefficiencies or diminishing returns in loan distribution. The study offers essential policy recommendations, such as expediting digital governance reforms, allocating financial resources to non-oil SMEs (SDG 8), aligning educational curricula with labor market demands, and implementing stricter environmental regulations to separate economic growth from emissions. Full article

The flowers of Panax ginseng C. A. Mey. (PG) and Panax notoginseng (Burkill) F. H. Chen ex C. H. Chow (PN) are morphologically indistinguishable after drying, leading to prevalent adulteration that compromises product quality and consumer safety. To address this issue, we developed a rapid, closed-tube molecular authentication method based on high-resolution melting (HRM) analysis. Species-specific primer pairs were designed to target the conserved ITS and rbcL-accD regions, with PNG-2 selected as the optimal candidate owing to its stable genotyping performance and moderate GC content. Our results established GC content, rather than amplicon length, as the primary determinant of the melting temperature (Tm). Notably, the experimentally measured Tm values were consistently 0.7–1.5 °C higher than theoretical predictions, a discrepancy attributable to the stabilizing effect of the saturated fluorescent dye. To ensure maximum diagnostic reliability, the HRM results were cross-validated through a three-tier system comprising ITS2 phylogenetic analysis, agarose gel electrophoresis, and Sanger sequencing. The practical utility and matrix robustness of the assay were further verified using a diversified validation cohort of 30 commercial samples, including 24 floral batches and 6 root-derived products (root slices and ultramicro powders). The HRM profiles demonstrated 100% concordance with DNA barcoding results, effectively identifying mislabeled products across different botanical matrices and processing forms. This methodology, which can be completed within 3 h, provides a significantly more cost-effective and rapid alternative to traditional sequencing-based methods for large-scale market surveillance and industrial quality control. Full article

Background: Root canal sealers may come into direct contact with periapical tissues, particularly in cases of apical extrusion, potentially influencing periapical healing and treatment outcomes. Cytotoxicity assessment represents a clinically relevant parameter when selecting endodontic sealers. However, evidence derived from in vitro studies remains heterogeneous and challenging to interpret from a clinical perspective. Therefore, the aim of this systematic review was to critically evaluate the in vitro cytotoxicity of all root canal sealers that have been commercially marketed over the years, excluding experimental materials, and to contextualize the findings in relation to clinically relevant experimental conditions. Methods: This systematic review was conducted according to PRISMA guidelines and preregistered on the Open Science Framework. PubMed, Scopus, and the Cochrane Library were searched up to 30 November 2025. In vitro studies evaluating the cytotoxicity of commercially available root canal sealers using validated cell viability or proliferation assays were included. Data extraction focused on sealer composition, setting condition, extraction protocols, exposure parameters, and cytotoxic outcomes. Due to marked methodological heterogeneity, a qualitative synthesis was performed. Results: Ninety-eight in vitro studies were included. All categories of root canal sealers demonstrated some degree of cytotoxicity, particularly when tested in freshly mixed conditions, at higher extract concentrations, or after prolonged exposure. Bioactive and calcium silicate-based sealers generally showed a more favorable cytotoxicity profile compared with conventional materials, especially after complete setting and at diluted concentrations, although cytotoxic effects were reported under specific experimental conditions. Resin-based sealers, including AH Plus, exhibited condition-dependent cytotoxicity, while zinc oxide–eugenol and glass ionomer sealers tended to display higher cytotoxic potential. Conclusions: In vitro cytotoxicity of root canal sealers varies according to material composition and experimental conditions. Bioactive sealers generally exhibit a more favorable biological profile, which may be clinically relevant in situations involving sealer extrusion or prolonged tissue contact. Standardized testing protocols and further translational studies are required to support evidence-based clinical material selection. Full article

Accurate grading of peach ripeness is essential for determining optimal harvest timing, postharvest storage potential, and market value. However, traditional methods are often inefficient and highly subjective. Meanwhile, existing deep learning models face challenges in balancing complexity with accuracy. To address this, this paper proposes a lightweight improved model named LightConvNeXt-FCS, which centers on a novel lightweight module named LightBlock. This module drastically reduces the parameter count and computational overhead. To preserve representational capacity, auxiliary structures—including attention enhancement, cross-stage connections, and multi-scale fusion—are incorporated. Experimental results show that the model requires only 2.75 M parameters and 624.23 M FLOPs, representing a 90.1% reduction in parameters and an 86.0% decrease in computational cost compared to ConvNeXt-Tiny, with the model size compressed to 9.9% of the original, while achieving an accuracy of 94.62%, slightly outperforming the original model. This approach effectively resolves the common trade-off between model complexity and accuracy. By achieving high accuracy with a lightweight architecture, it provides a more feasible solution for deploying rapid and intelligent fruit ripeness grading systems. Full article

Workplace weight stigma is a form of discrimination affecting equality, health, and careers, yet occupational research remains fragmented. This gender-sensitive systematic review synthesizes evidence on workplace weight stigma among adult workers and job applicants since 2000. Following PRISMA procedures, we searched psychological, medical, sociological, and economic databases, identifying 25 included studies examining work outcomes. The corpus includes experimental vignette and correspondence studies, surveys, and qualitative designs, predominantly from high-income Western countries. Higher body weight is consistently associated with disadvantages across the employment life cycle: reduced callbacks and hiring, lower wages and wage growth, fewer promotions, and negative performance evaluations. Penalties are systematically stronger for women; intersectional analyses remain rare. Weight-based teasing, unfair treatment, and stereotype threat are linked to poorer self-rated health, psychological distress, burnout, reduced work ability, lower job satisfaction and commitment, and stronger turnover intentions. Organizational-level evidence is indirect but suggests detrimental effects on engagement and citizenship behaviors. Findings support conceptualizing workplace weight stigma as both a psychosocial hazard and a structural driver of labor-market inequality, underscoring the need for size-inclusive HR practices, leadership, and occupational risk-prevention policies. Full article