Search Results (2,318)

Biological invasions provide a unique window into community assembly. While classic theory predicts that native species must differentiate their niches to coexist with an invader, the actual outcomes under intense pressure are complex. Our study examines community reassembly under extreme pressure from the invasive ant Solenopsis invicta. We found that while native species do differentiate themselves from the invader, the overwhelming competition constrains this process, forcing survivors into a narrow, shared functional space. This constrained niche differentiation produces a pattern of community-level functional convergence, a process where functionally dissimilar communities become more similar under intense environmental filtering, as survivors are forced into a narrow, shared niche space. The capacity for these rapid, adaptive niche shifts is rooted in intraspecific trait variation (ITV). We also identified a dynamic feedback loop through density-dependent phenotypic plasticity in the invader. By showing how the foundational process of niche differentiation leads to a convergent outcome under extreme pressure, our work clarifies the rules of community assembly in an increasingly invaded world. Full article

Background: Tendon structure is related to the magnitude of load and its management, rather than directly predicting pain incidence. Although pain symptoms frequently persist until they severely impact performance, function, and tendon structure, professional basketball players often manage patellar tendon pain alongside high training and competition loads. The aim of this study was to investigate patellar tendon structural adaptations over 8 months of training and competition in professional female and male basketball players. Methods: The primary outcome of this study was defined as the change in the percentage of echo-type II structure from baseline to 8 months. A prospective cohort study was conducted where 43 professional basketball players (20 male, 23 female) were followed during 8 months of training and competition. A bilateral patellar tendon ultrasound tissue characterization (UTC) scan was conducted, and jumping leg (jumping/non-jumping), presence of pain (yes/no), and exposure time (hours) were recorded at baseline, and at 4 and 8 months in the competitive season. Results: The mid-tendon exhibited negative adaptations (decreased alignment), represented by echo-type II, after 8 months of competition (left; p = 0.001; ES = 2.16; right; p = 0.001; ES = −1.35). Positive adaptations (increased alignment], were observed in echo-types III (p = 0.004; ES = −1.04) and IV (p = 0.001; ES = −0.15), after 4 and 8 months. Tendon structure showed differences between female and male professional basketball players throughout the 8 months (echo-type II; p = 0.00; ES = 0.34). Conclusions: The study demonstrated that the tendon structure undergoes significant adaptations, supporting the concept that the patellar tendon adapts to compensate for areas of structure disorganization. Female professional basketball players appeared to maintain a more organized tendon structure than their male counterparts. Jumping leg and presence of pain did not show significant differences in tendon structure over the study period. This research has significant implications for elite sport, as a better understanding of tendon load capacity throughout a competitive season is needed. Full article

Accurate branch prediction is crucial for achieving high instruction throughput and minimizing control hazards in modern pipelines. This paper presents a novel LXOR (Local eXclusive-OR) branch predictor, which enhances prediction accuracy while reducing hardware complexity and memory usage. Unlike traditional predictors (GAg, GAp, PAg, PAp, Gshare, Gselect) that rely on large Pattern History Tables (PHTs) or intricate global/local history combinations, the LXOR predictor employs complemented local history and XOR-based indexing, optimizing table access and reducing aliasing. Implemented and evaluated using the MARSS-RISCV simulator on a 64-bit in-order RISC-V core, the LXOR’s performance was compared against traditional predictors using Coremark and SPEC CPU2017 benchmarks. The LXOR consistently achieved competitive results, with a prediction accuracy of up to 83.92%, lower misprediction rates, and instruction flushes as low as 5.83%. It also attained an IPC rate of up to 0.83, all while maintaining a compact memory footprint of approximately 2 KB, significantly smaller than current alternatives. These findings demonstrate that the LXOR predictor not only matches the performance of more complex predictors but does so with less memory and logic overhead, making it ideal for embedded systems, low-power RISC-V processors, and resource-constrained IoT and edge devices. By balancing prediction accuracy with simplicity, the LXOR offers a scalable and cost-effective solution for next-generation microprocessors. Full article

Accurate estimation of lithium-ion battery (LIB) state of health (SOH) underpins safe operation, predictive maintenance, and lifetime-aware energy management. Despite recent advances in machine learning (ML), systematic benchmarking across heterogeneous real-world cells remains limited, often confounded by data leakage and inconsistent validation. Here, we establish a leakage-averse, cross-battery evaluation framework encompassing 32 commercial LIBs (B5–B56) spanning diverse cycling histories and temperatures (≈4 °C, 24 °C, 43 °C). Models ranging from classical regressors to ensemble trees and deep sequence architectures were assessed under blocked 5-fold GroupKFold splits using RMSE, MAE, R² with confidence intervals, and inference latency. The results reveal distinct stratification among model families. Sequence-based architectures—CNN–LSTM, GRU, and LSTM—consistently achieved the highest accuracy (mean RMSE ≈ 0.006; per-cell R² up to 0.996), demonstrating strong generalization across regimes. Gradient-boosted ensembles such as LightGBM and CatBoost delivered competitive mid-tier accuracy (RMSE ≈ 0.012–0.015) yet unrivaled computational efficiency (≈0.001–0.003 ms), confirming their suitability for embedded applications. Transformer-based hybrids underperformed, while approximately one-third of cells exhibited elevated errors linked to noise or regime shifts, underscoring the necessity of rigorous evaluation design. Collectively, these findings establish clear deployment guidelines: CNN–LSTM and GRU are recommended where robustness and accuracy are paramount (cloud and edge analytics), while LightGBM and CatBoost offer optimal latency–efficiency trade-offs for embedded controllers. Beyond model choice, the study highlights data curation and leakage-averse validation as critical enablers for transferable and reliable SOH estimation. This benchmarking framework provides a robust foundation for future integration of ML models into real-world battery management systems. Full article

Background: Patients with chronic lymphocytic leukemia (CLL) who were not receiving treatment were included in this experimental prospective correlation study. We aimed to elucidate the complex relationship between smudge cells, surface CD20, and soluble CD20 in CLL patients. Methods: We created blood smears from blood samples collected from our patients using a manual technique consistently performed by the same technician. The May–Grunwald Giemsa dye was used to stain all of the slides. The B-cell phenotypic was analyzed using the FacsCanto II flow cytometer (Becton Dickinson, CA, USA) at the time of diagnosis. Competitive Enzyme-Linked Immunoassay (ELISA) was used to quantitatively assess the amounts of soluble CD20/MS4A1. Results: The percentage of smudge cells and soluble CD20 antigen levels were shown to be significantly inversely correlated, suggesting a considerable link (correlation coefficient (r) = −0.51, p = 0.006). Similarly, a significant inverse relationship (r = −0.36, p = 0.04) was found by the Spearman correlation test between the smudge cell ratio and CD20 median fluorescence intensity (MFI) on cell surfaces. Soluble CD20/MS4A1 and surface CD20 MFI were shown to have a weakly positive association that was almost statistically significant (Spearman’s rho = 0.34, p = 0.064). With a sensitivity of 69% and specificity of 86%, we discovered that a cut-off value of 2.2 ng/dL for soluble CD20 predicted higher smudge cells (area under the curve (95% confidence interval (CI)): 0.75 (0.57 to 0.93), p = 0.021). Conclusions: We found a significant inverse association between smudge cells and both surface CD20 and soluble CD20/MS4A1 in our study examining the correlation between smudge cells, soluble CD20, and CD20/MS4A1 in CLL patients. Our findings indicate that soluble CD20 may contribute to understanding the pathophysiology of smudge cells and could be further investigated as a potential prognostic marker in CLL. Full article

Dynamic technological changes and the variability of market requirements pose significant challenges for modern manufacturing companies in the effective development and implementation of new technological solutions. The aim of the research was to develop an integrated approach covering all key stages of implementation—from formulating technological solutions, through selecting and evaluating variants, to preparing and managing production processes—under the conditions of a medium-sized manufacturing company specializing in the batch production of steel constructions. The analysis was based on an interdisciplinary approach, combining methods of creative design of new technological solutions, including Blue Ocean Strategy, value proposition design, and QFD methodology, with analytical approaches that include multi-criteria evaluation of solution variants, technical preparation of production, as well as the organization and management of production processes in modified organizational conditions. This approach enabled a comprehensive assessment of the developed solutions, taking into account both their operational potential and practical feasibility in realistic implementation conditions, through the use of case studies and simulations to validate the results. The results of the research indicate that integrating methods for creating new solutions with analytical assessment and simulation tools leads to a more precise and data-driven approach to process design, enabling better decision-making based on thorough analysis and predictive modeling. Furthermore, this approach allows for a significant reduction in the risk of implementation failure through early identification of potential problems. The conclusion of the study confirms that a comprehensive and interdisciplinary approach to the implementation of new technologies ensures better alignment with customer demands, reduces production downtime, and enhances product optimization and resource utilization, which are critical factors in building a sustainable competitive advantage for manufacturing companies. The proposed approach enables more deliberate design and organization of manufacturing processes, supporting their flexible adaptation to changing market and technological conditions. Full article

Accurate traffic flow prediction is pivotal for intelligent transportation systems; yet, existing spatial-temporal graph neural networks (STGNNs) struggle to jointly capture the long-term structural stability, short-term dynamics, and multi-scale temporal patterns of road networks. To address these shortcomings, we propose FISTGCN, a Frequency-Aware Interactive Spatial-Temporal Graph Convolutional Network. FISTGCN enriches raw traffic flow features with learnable spatial and temporal embeddings, thereby providing comprehensive spatial-temporal representations for subsequent modeling. Specifically, it utilizes an interactive dynamic graph convolutional block that generates a time-evolving fused adjacency matrix by combining adaptive and dynamic adjacency matrices. It then applies dual sparse graph convolutions with cross-scale interactions to capture multi-scale spatial dependencies. The gated spectral block projects the input features into the frequency domain and adaptively separates low- and high-frequency components using a learnable threshold. It then employs learnable filters to extract features from different frequency bands and adopts a gating mechanism to adaptively fuse low- and high-frequency information, thereby dynamically highlighting short-term fluctuations or long-term trends. Extensive experiments on four benchmark datasets demonstrate that FISTGCN delivers state-of-the-art predictive accuracy while maintaining competitive computational efficiency. Full article

Molecular simulation is central to modern drug discovery but is often limited by high computational cost and the complexity of molecular interactions. Deep-learning drug–target interaction (DTI) prediction can accelerate screening; however, many models underuse the local functional structure features—binding motifs, reactive groups, and residue-level fragments—that drive recognition. We present LoF-DTI, a framework that explicitly represents and couples such local features. Drugs are converted from SMILES into molecular graphs and targets from sequences into feature representations. On the drug side, a Jumping Knowledge (JK) enhanced Graph Isomorphism Network (GIN) extracts atom- and neighborhood-level patterns; on the target side, residual CNN blocks with progressively enlarged receptive fields, augmented by N-mer substructural statistics, capture multi-scale local motifs. A Gated Cross-Attention (GCA) module then performs atom-to-residue interaction learning, highlighting decisive local pairs and providing token-level interpretability through attention scores. By prioritizing locality during both encoding and interaction, LoF-DTI delivers competitive results across multiple benchmarks and improves early retrieval relevant to virtual screening. Case analyses show that the model recovers known functional binding sites, suggesting strong potential to provide mechanism-aware guidance for molecular simulation and to streamline the drug design pipeline. Full article

Website performance directly impacts user experience, trust, and competitiveness. While numerous studies have proposed evaluation methods, there is still no comprehensive synthesis that integrates performance metrics with predictive models. This study conducts a systematic literature review (SLR) following the PRISMA framework across seven academic databases (2010–2024). From 6657 initial records, 30 high-quality studies were included after rigorous screening and quality assessment. In addition, 59 website performance metrics were identified and validated through an expert survey, resulting in 16 core indicators. The review highlights a dominant reliance on traditional evaluation metrics (e.g., Load Time, Page Size, Response Time) and reveals limited adoption of machine learning and deep learning approaches. Most existing studies focus on e-government and educational websites, with little attention to e-commerce, healthcare, and industry domains. Furthermore, the geographic distribution of research remains uneven, with a concentration in Asia and limited contributions from North America and Africa. This study contributes by (i) consolidating and validating a set of 16 critical performance metrics, (ii) critically analyzing current methodologies, and (iii) identifying gaps in domain coverage and intelligent prediction models. Future research should prioritize cross-domain benchmarks, integrate machine learning for scalable predictions, and address the lack of standardized evaluation protocols. Full article

This article takes the perspective of Hydrogen Load Aggregator (HLA) to optimize the declaration strategy of peak shaving market, improve the flexible regulation capability of power system and HLA economy as the research objectives, and proposes an optimization strategy method for HLA to participate in peak shaving market. Firstly, an improved Convolutional Neural Networks–Long Short-Term Memory (CNN-LSTM) time series prediction model is developed to address peak shaving demand uncertainty. Secondly, a bidding strategy model incorporating dynamic pricing is constructed by comprehensively considering electrolyzer regulation costs, market supply–demand relationships, and system constraints. Thirdly, a market clearing model for peak shaving markets with HLA participation is designed through analysis of capacity contribution and marginal costs among different regulation resources. Finally, the capacity allocation model is designed with the goal of minimizing the total cost of peak shaving among various stakeholders within HLA, and the capacity won by HLA in the peak shaving market is reasonably allocated. Simulations conducted on a Python3.12-based experimental platform demonstrate the following: the improved CNN-LSTM model exhibits strong adaptability and robustness, the bidding model effectively enhances HLA market competitiveness, and the clearing model reduces system operator costs by 5.64%. Full article

Customer loyalty is crucial for (while fueled by) business sustainability. Loyal customers advocate for a company’s offer and sustainable practices, while their steady support generates stable revenue stream, lower acquisition costs, and predictable cash flows that enable long-term business viability. Such a stable revenue stream is especially critical in periods of intense competition or macroeconomic disruption (e.g., COVID-19 pandemic) which profoundly influenced consumer behavior. In this context, the purpose of the current paper is to test the BG/NBD prediction model for its potential validation as a practical tool in estimating the buying behavior of customers of a self-service car washing company before and within the COVID-19 pandemic period. To achieving this, transaction data of the company’s customers was retrieved from the company’s internal information system and used as input for BG/NBD model. The model proved its effectiveness in estimating the total number of repeated transactions for the year 2020 based on the 2019 data at total customer base and at loyal customer level. Loyal customers were considered from the behavioral loyalty perspective only and defined as customers which used the company’s services at least once in both years. In the estimation of the repeated transactions frequencies, the model’s prediction accuracy increases with higher frequencies of loyal customers. Full article

Fluoroscopy (real-time X-ray) images are used for monitoring minimally invasive coronary angioplasty operations such as stent placement. During these operations, a thin wire called a guidewire is used to guide different tools, such as a stent or a balloon, in order to repair the vessels. However, fluoroscopy images are noisy, and the guidewire is very thin, practically invisible in many places, making its localization very difficult. Guidewire segmentation is the task of finding the guidewire pixels, while guidewire localization is the higher-level task aimed at finding a parameterized curve describing the guidewire points. This paper presents a method for guidewire localization that starts from a guidewire segmentation, from which it extracts a number of initial curves as pixel chains and uses a novel perceptual grouping method to merge these initial curves into a small number of curves. The paper also introduces a novel guidewire segmentation method that uses a residual network (ResNet) as a feature extractor and predicts a coarse segmentation that is refined only in promising locations to a fine segmentation. Experiments on two challenging datasets, one with 871 frames and one with 23,449 frames, show that the method obtains results competitive with existing segmentation methods such as Res-UNet and nnU-Net, while having no skip connections and a faster inference time. Full article

The efficient and non-destructive inspection of the chlorophyll content of grey jujube leaf is of great significance for its growth surveillance and nutritional diagnosis. Near-infrared spectroscopy combined with chemometric methods provides an effective approach to achieve this goal. This study took grey jujube leaves as the research object, systematically collected near-infrared spectral data in the range of 4000–10,000 cm⁻¹, and simultaneously measured their soil and plant analyzer development (SPAD) value as a reference index for chlorophyll content. Through various pretreatment and their combination methods on the original spectrum—smooth, standard normal variable transformation (SNV), first derivative (FD), second derivative (SD), smooth + first derivative (Smooth + FD), smooth + second derivative (Smooth + SD), standard normal variable transformation + first derivative (SNV + FD), standard normal variable transformation + second derivative (SNV + SD)—the effects of different methods on the quality of the spectrum and its correlation with SPAD value were compared. The competitive adaptive reweighted sampling algorithm (CARS) was adopted to extract the characteristic wavelength, aiming to reduce data dimensionality and optimize model input. Both BP neural network and RBF neural network prediction models were established, and the model performance under different training functions was compared. The results indicate that after Smooth + FD pretreatment, followed by CARS screening of the characteristic wavelength, the BP neural network model trained using the LBFGS algorithm demonstrated the best performance, with its coefficient of determination (R²) of 0.87 (training set) and 0.85 (validation set), root mean square error (RMSE) of 1.36 (training set) and 1.35 (validation set), and residual prediction deviation (RPD) of 2.81 (training set) and 2.56 (validation set) showing good prediction accuracy and robustness. Research indicates that by combining near-infrared spectroscopy with feature extraction and machine learning methods, the rapid and non-destructive inspection of the grey jujube leaf SPAD value can be achieved, providing reliable technical support for the real-time monitoring of the nutritional status of jujube trees. Full article

Semiconductor wafer manufacturing is one of the most complex and data-intensive industrial processes, comprising 500–1000 tightly interdependent steps, each requiring nanometer-level precision. As device nodes approach 3 nm and beyond, even minor deviations in parameters such as oxide thickness or critical dimensions can lead to catastrophic yield loss, challenging traditional physics-based control methods. In response, the industry has increasingly adopted regression analysis and predictive modeling as essential analytical frameworks. Classical regression, long used to support design of experiments (DOE), process optimization, and yield analysis, has evolved to enable multivariate modeling, virtual metrology, and fault detection. Predictive modeling extends these capabilities through machine learning and AI, leveraging massive sensor and metrology data streams for real-time process monitoring, yield forecasting, and predictive maintenance. These data-driven tools are now tightly integrated into advanced process control (APC), digital twins, and automated decision-making systems, transforming fabs into agile, intelligent manufacturing environments. This review synthesizes foundational and emerging methods, industry applications, and case studies, emphasizing their role in advancing Industry 4.0 initiatives. Future directions include hybrid physics–ML models, explainable AI, and autonomous manufacturing. Together, regression and predictive modeling provide semiconductor fabs with a robust ecosystem for optimizing performance, minimizing costs, and accelerating innovation in an increasingly competitive, high-stakes industry. Full article

Financial news has a significant impact on investor sentiment and short-term stock price trends. While many studies have applied natural language processing (NLP) techniques to financial forecasting, most have focused on single tasks or English corpora, with limited research in non-English language contexts such as Taiwan. This study develops a joint framework to perform sentiment classification and short-term stock price prediction using Chinese financial news from Taiwan’s top 50 listed companies. Five types of word embeddings—one-hot, TF-IDF, CBOW, skip-gram, and BERT—are systematically compared across 17 traditional, deep, and Transformer models, as well as a large language model (LLaMA3) fully fine-tuned on the Chinese financial texts. To ensure annotation quality, sentiment labels were manually assigned by annotators with finance backgrounds and validated through a double-checking process. Experimental results show that a CNN using skip-gram embeddings achieves the strongest performance among deep learning models, while LLaMA3 yields the highest overall F1-score for sentiment classification. For regression, LSTM consistently provides the most reliable predictive power across different volatility groups, with Bayesian Linear Regression remaining competitive for low-volatility firms. LLaMA3 is the only Transformer-based model to achieve a positive $R^2$ under high-volatility conditions. Furthermore, forecasting accuracy is higher for the five-day horizon than for the fifteen-day horizon, underscoring the increasing difficulty of medium-term forecasting. These findings confirm that financial news provides valuable predictive signals for emerging markets and that short-term sentiment-informed forecasts enhance real-time investment decisions. Full article