Search Results (32,260)

Environmental protection has become one of the key challenges of our time. This has led to an increase in pro-environmental activities in the field of cosmetics and household chemicals, where manufacturers are increasingly trying to meet the expectations of consumers who are aware of the potential risks associated with the production of cosmetics and household chemistry products. This is one of the most important challenges of today’s industry, given that some of the raw materials still commonly used, such as surfactants, may be toxic to aquatic organisms. Many companies are choosing to use natural raw materials that have satisfactory performance properties but are also environmentally friendly. In addition, modern products are also characterized by reduced consumption of water, resources, and energy in production processes. These measures reduce the carbon footprint and reduce the amount of plastic packaging required. In the present study, seven formulations of environmentally friendly car shampoo concentrates were developed, based entirely on mixtures of bio-based surfactants. The developed formulations were tested for application on the car body surface, allowing the selection of the two best products. For these selected formulations, an in-depth physicochemical analysis was carried out, including pH, density, and viscosity measurements. Comparison of the results with commercial products available on the market was also performed. Additionally, using the multiple light scattering method, the foamability and foam stability were determined for the car shampoos developed. The results obtained indicate the very high application potential of the products under study, which combine high performance and environmental concerns. Full article

Adversarial attacks in Natural Language Processing (NLP) present a critical challenge, particularly in sentiment analysis, where subtle input modifications can significantly alter model predictions. In search of more robust defenses against adversarial attacks on sentimental analysis, this research work introduces two novel defense mechanisms: the Lexicon-Based Random Substitute Model (LRSM) and the Word-Variant Voting Model (WVVM). LRSM employs randomized substitutions from a dataset-specific lexicon to generate diverse input variations, disrupting adversarial strategies by introducing unpredictability. Unlike traditional defenses requiring synonym dictionaries or precomputed semantic relationships, LRSM directly substitutes words with random lexicon alternatives, reducing overhead while maintaining robustness. Notably, LRSM not only neutralizes adversarial perturbations but occasionally surpasses the original accuracy by correcting inherent model misclassifications. Building on LRSM, WVVM integrates LRSM, Frequency-Guided Word Substitution (FGWS), and Synonym Random Substitution and Voting (RS&V) in an ensemble framework that adaptively combines their outputs. Logistic Regression (LR) emerged as the optimal ensemble configuration, leveraging its regularization parameters to balance the contributions of individual defenses. WVVM consistently outperformed standalone defenses, demonstrating superior restored accuracy and F1 scores across adversarial scenarios. The proposed defenses were evaluated on two well-known sentiment analysis benchmarks: the IMDB Sentiment Dataset and the Yelp Polarity Dataset. The IMDB dataset, comprising 50,000 labeled movie reviews, and the Yelp Polarity dataset, containing labeled business reviews, provided diverse linguistic challenges for assessing adversarial robustness. Both datasets were tested using 4000 adversarial examples generated by established attacks, including Probability Weighted Word Saliency, TextFooler, and BERT-based Adversarial Examples. WVVM and LRSM demonstrated superior performance in restoring accuracy and F1 scores across both datasets, with WVVM excelling through its ensemble learning framework. LRSM improved restored accuracy from 75.66% to 83.7% when compared to the second-best individual model, RS&V, while the Support Vector Classifier WVVM variation further improved restored accuracy to 93.17%. Logistic Regression WVVM achieved an F1 score of 86.26% compared to 76.80% for RS&V. These findings establish LRSM and WVVM as robust frameworks for defending against adversarial text attacks in sentiment analysis. Full article

Objective: We evaluated the feasibility of a machine-learning (ML) model based on clinical features and radiomics from [¹⁸F]FDG PET/CT images to differentiate between infected and non-infected intracavitary vascular grafts and endografts (iVGEI). Methods: Three ML models were developed: one based on pre-treatment criteria to diagnose a vascular graft infection (“MAGIC-light features”), another using radiomics features from diagnostic [¹⁸F]FDG-PET scans, and a third combining both datasets. The training set included 92 patients (72 iVGEI-positive, 20 iVGEI-negative), and the external test set included 20 iVGEI-positive and 12 iVGEI-negative patients. The abdominal aorta and iliac arteries in the PET/CT scans were automatically segmented using SEQUOIA and TotalSegmentator and manually adjusted, extracting 96 radiomics features. The best-performing models for the MAGIC-light features and PET-radiomics features were selected from 343 unique models. Most relevant features were combined to test three final models using ROC analysis, accuracy, sensitivity, and specificity. Results: The combined model achieved the highest AUC in the test set (mean ± SD: 0.91 ± 0.02) compared with the MAGIC-light-only model (0.85 ± 0.06) and the PET-radiomics model (0.73 ± 0.03). The combined model also achieved a higher accuracy (0.91 vs. 0.82) than the diagnosis based on all the MAGIC criteria and a comparable sensitivity and specificity (0.70 and 1.00 vs. 0.76 and 0.92, respectively) while providing diagnostic information at the initial presentation. The AUC for the combined model was significantly higher than the PET-radiomics model (p = 0.02 in the bootstrap test), while other comparisons were not statistically significant. Conclusions: This study demonstrated the potential of ML models in supporting diagnostic decision making for iVGEI. A combined model using pre-treatment clinical features and PET-radiomics features showed high diagnostic performance and specificity, potentially reducing overtreatment and enhancing patient outcomes. Full article

The high cost of hydrogen production is the primary factor limiting the development of the hydrogen energy industry chain. Additionally, due to the inefficiency of hydrogen production by water electrolysis technology, the development of high-performance catalysts is an effective means of producing low-cost hydrogen. In water electrolysis technology, the electrocatalytic activity of the electrode affects the kinetics of the oxygen evolution reaction (OER) and the hydrogen evolution rate. This study utilizes the liquid phase co-precipitation method to synthesize three types of Prussian blue analog (PBA) electrocatalytic materials: Fe/PBA(Fe₄[Fe(CN)₆]₃), Fe-Mn/PBA((Fe, Mn)₃[Fe(CN)₆]₂·nH₂O), and Fe-Mn-Co/PBA((Mn, Co, Fe)₃^II[Fe^III(CN)₆]₂·nH₂O). X-ray diffraction (XRD) and scanning electron microscopy (SEM) analyses show that Fe-Mn-Co/PBA has a smaller particle size and higher crystallinity, and its grain boundary defects provide more active sites for electrochemical reactions. The electrochemical test shows that Fe-Mn-Co/PBA exhibits the best electrochemical performance. The overpotential of the oxygen evolution reaction (OER) under 1 M alkaline electrolyte at 10/50 mA·cm⁻² is 270/350 mV, with a Tafel slope of 48 mV·dec⁻¹, and stable electrocatalytic activity is maintained at 5 mA·cm⁻². All of these are attributed to the synergistic effect of Fe, Mn, and Co metal ions, grain refinement, and the generation of grain boundary defects and internal stresses. Full article

Accurately predicting the next location of urban individuals is a central issue in human mobility research. Human mobility exhibits diverse patterns, requiring the integration of spatiotemporal contexts for location prediction. In this context, multi-view learning has become a prominent method in location prediction. Despite notable advances, current methods still face challenges in effectively capturing non-spatial proximity of regional preferences, complex temporal periodicity, and the ambiguity of location semantics. To address these challenges, we propose a representation-enhanced multi-view learning network (ReMVL-Net) for location prediction. Specifically, we propose a community-enhanced spatial representation that transcends geographic proximity to capture latent mobility patterns. In addition, we introduce a multi-granular enhanced temporal representation to model the multi-level periodicity of human mobility and design a rule-based semantic recognition method to enrich location semantics. We evaluate the proposed model using mobile phone data from Fuzhou. Experimental results show a 2.94% improvement in prediction accuracy over the best-performing baseline. Further analysis reveals that community space plays a key role in narrowing the candidate location set. Moreover, we observe that prediction difficulty is strongly influenced by individual travel behaviors, with more regular activity patterns being easier to predict. Full article

Catalytic upgrading of pyrolysis oils is an important step for producing replacement hydrocarbon-rich liquid biofuels from biomass and can help to advance pyrolysis technology. Catalysts play a pivotal role in influencing the selectivity of chemical reactions leading to the formation of main compounds in the final upgraded liquid products. The present work involved a systematic study of solvent-free catalytic reactions of cyclohexanone in the presence of hydrogen gas at 160 °C for 3 h in a batch reactor. Cyclohexanone can be produced from biomass through the selective hydrogenation of lignin-derived phenolics. Three types of catalysts comprising undoped NbOPO₄, 10 wt% NiO/NbOPO₄, and 30 wt% NiO/NbOPO₄ were studied. Undoped NbOPO₄ promoted both aldol condensation and the dehydration of cyclohexanol, producing fused ring aromatic hydrocarbons and hard char. With 30 wt% NiO/NbOPO₄, extensive competitive hydrogenation of cyclohexanone to cyclohexanol was observed, along with the formation of C₆ cyclic hydrocarbons. When compared to NbOPO₄ and 30 wt% NiO/NbOPO₄, the use of 10 wt% NiO/NbOPO₄ produced superior selectivity towards bi-cycloalkanones (i.e., C₁₂) at cyclohexanone conversion of 66.8 ± 1.82%. Overall, the 10 wt% NiO/NbOPO₄ catalyst exhibited the best performance towards the production of precursor compounds that can be further hydrodeoxygenated into energy-dense aviation fuel hydrocarbons. Hence, the presence and loading of NiO was able to tune the activity and selectivity of NbOPO₄, thereby influencing the final products obtained from the same cyclohexanone feedstock. This study underscores the potential of lignin-derived pyrolysis oils as important renewable feedstocks for producing replacement hydrocarbon solvents or feedstocks and high-density sustainable liquid hydrocarbon fuels via sequential and selective catalytic upgrading. Full article

The aim of this study was to evaluate the impact of autoclaving on the dimensional stability of surgical guides (SGs) for aesthetic crown lengthening (ACL) using different resins/printing methods. Fifty SGs for ACL were printed using five different resin/printer combinations (FL, SR, ND, KS and VC). All the SGs were scanned before (T0) and after (T1) sterilization. Autoclaving was conducted at 134 °C during 4 min. The STL files of each SG at T0 and T1 were compared with the original design (TR). Dimensional stability was measured using trueness and precision. Deviations from TR to T1 were calculated in the three space axes and by measuring the area between three reference landmarks. At T0, the FL group showed the best trueness and precision, while the SR group performed significantly worse than the other groups. At T1, all the groups except VC exhibited significant dimensional alterations compared with T0. Also, VC showed the best trueness and precision values. All the groups had a significant deviation in at least one space axis, while only the SR group exhibited significant variations from T1 to TR in the area between the reference landmarks. Most of the evaluated resin/3D printer combinations suffered significant dimensional alterations after autoclaving. Full article

Accurate melt pool geometry prediction is essential for ensuring quality and reliability in Laser Powder Bed Fusion (L-PBF). However, small experimental datasets and limited physical interpretability often restrict the effectiveness of traditional machine learning (ML) models. This study proposes a hybrid framework that integrates an explicit thermal model with ML algorithms to improve prediction under sparse data conditions. The explicit model—calibrated for variable penetration depth and absorptivity—generates synthetic melt pool data, augmenting 36 experimental samples across conduction, transition, and keyhole regimes for 316 L stainless steel. Three ML methods—Multilayer Perceptron (MLP), Random Forest, and XGBoost—are trained using fivefold cross-validation. The hybrid approach significantly improves prediction accuracy, especially in unstable transition regions (D/W ≈ 0.5–1.2), where morphological fluctuations hinder experimental sampling. The best-performing model (MLP) achieves R² > 0.98, with notable reductions in MAE and RMSE. The results highlight the benefit of incorporating physically consistent, nonlinearly distributed synthetic data to enhance generalization and robustness. This physics-augmented learning strategy not only demonstrates scientific novelty by integrating mechanistic modeling into data-driven learning, but also provides a scalable solution for intelligent process optimization, in situ monitoring, and digital twin development in metal additive manufacturing. Full article

Agriculture is the backbone of Punjab’s economy, and with much of India’s population dependent on agriculture, the requirement for accurate and timely monitoring of land has become even more crucial. Blending remote sensing with state-of-the-art machine learning algorithms enables the detailed classification of agricultural lands through thematic mapping, which is critical for crop monitoring, land management, and sustainable development. Here, a Hyper-tuned Deep Neural Network (Hy-DNN) model was created and used for land use and land cover (LULC) classification into four classes: agricultural land, vegetation, water bodies, and built-up areas. The technique made use of multispectral data from Sentinel-2 and Landsat-8, processed on the Google Earth Engine (GEE) platform. To measure classification performance, Hy-DNN was contrasted with traditional classifiers—Convolutional Neural Network (CNN), Random Forest (RF), Classification and Regression Tree (CART), Minimum Distance Classifier (MDC), and Naive Bayes (NB)—using performance metrics including producer’s and consumer’s accuracy, Kappa coefficient, and overall accuracy. Hy-DNN performed the best, with overall accuracy being 97.60% using Sentinel-2 and 91.10% using Landsat-8, outperforming all base models. These results further highlight the superiority of the optimised Hy-DNN in agricultural land mapping and its potential use in crop health monitoring, disease diagnosis, and strategic agricultural planning. Full article

Agricultural production in southern Angola faces challenges due to unsustainable practices, including inefficient use of water, fertilizers, and machinery, resulting in low yields and environmental degradation. Therefore, clear and measurable indicators are needed to guide farmers toward more sustainable practices. The scientific literature insufficiently addresses this issue, leaving a significant gap in the evaluation of key performance indicators (KPIs) that can guide good agricultural practices (GAPs) adapted to the context of southern Angola, with the goal of promoting a more resilient and sustainable agricultural sector. So, the objective of this study is to identify and assess KPIs capable of supporting the selection of GAPs suitable for maize, potato, and tomato cultivation in the context of southern Angolan agriculture. A systematic literature review (SLR) was conducted, screening 2720 articles and selecting 14 studies that met defined inclusion criteria. Five KPIs were identified as the most relevant: gross margin, net profit, water use efficiency, nitrogen use efficiency, and machine energy. These indicators were analyzed and standardized to evaluate their contribution to sustainability across different GAPs. Results show that organic fertilizers are the most sustainable option for maize, drip irrigation for potatoes, and crop rotation for tomatoes in southern Angola because of their efficiency in low-resource environments. A clear, simple, and effective representation of the KPIs was developed to be useful in communicating to farmers and policy makers on the selection of the best GAPs in the cultivation of different crops. The study proposes a validated KPI-based methodology for assessing sustainable agricultural practices in developing regions such as southern Angola, aiming to lead to greater self-sufficiency and economic stability in this sector. Full article

In this work, the biosorption potential of Spirulina sp. as an effective and eco-friendly biosorbent for the removal of Ni(II) and Pb(II) ions from aqueous solutions was investigated. Detailed characterization of the biosorbent was carried out, including surface morphology, chemical composition, particle size, zeta potential, crystallinity, zero-point charge, and functional group analysis. Batch tests were performed to determine the kinetic constants and adsorption equilibrium of the studied ions. The adsorption behavior of Spirulina sp. was described using six adsorption isotherms. The best fit was obtained for the Redlich-Peterson and Langmuir isotherms, indicating that monolayer adsorption occurred. The maximum biosorption capacities for Ni(II) and Pb(II) were 20.8 mg·g⁻¹ and 93.5 mg·g⁻¹, respectively, using a biosorbent dose of 10 g·L⁻¹, initial metal concentrations ranging from 50 to 5000 mg·L⁻¹, at pH 6, 20 °C, and a contact time of 120 min. Low values of the mean free energy of adsorption (E) in the Dubinin–Radushkevich and Temkin model (0.3 and 0.1 kJ·mol⁻¹ for Pb(II) and 0.35 and 0.23 kJ·mol⁻¹ for Ni(II)) indicate the dominance of physical processes in the ion binding mechanism. The adsorption of Pb(II) ions was more effective than that of Ni(II) ions across the entire range of tested concentrations. At low initial concentrations, the removal of Pb(II) reached 94%, while for Ni(II) it was 80%. Full article

Ecotribology focuses on both saving energy resources and reducing environmental pollution. Considering environmental concerns, water-based nanolubricants have gained significant attention over conventional oil-based ones. Non-ecotoxic and highly environmentally friendly nanoadditives were chosen for nanolubricant synthesis, especially considering their use at elevated temperatures. In this study, hexagonal boron nitride nanosheets (hBNNSs) and titanium dioxide nanoparticles (TiO₂ NPs) were used to prepare water-based lubricants with glycerol and surfactant sodium dodecyl benzene sulfonate (SDBS) in water under ultrasonication. An Rtec ball-on-disk tribometer was used to investigate the tribological performance of the synthesised water-based lubricants containing different nano-hBN/TiO₂ concentrations, with dry and water conditions used as benchmarks. The results indicated that the water-based nanolubricant containing 0.5 wt% hBN and 0.5 wt% TiO₂ exhibited the best tribological performance at both ambient (25 °C) and elevated (500 °C) temperatures. This optimal concentration leads to a reduction in the coefficient of friction (COF) by 72.9% and 37.5%, wear of disk by 62.5% and 49%, and wear of ball by 74% and 69% at ambient and elevated temperatures, respectively, compared to that of distilled water. Lubrication mechanisms were attributed to the rolling, mending, tribofilm, solid layer formation, and synergistic effects of hBNNSs and TiO₂ NPs. Full article

In order to increase the effectiveness of model training, data reduction is essential to data-centric Artificial Intelligence (AI). It achieves this by locating the most instructive examples in massive datasets. To increase data quality and training efficiency, the main difficulty is choosing the best examples rather than the complete datasets. In this paper, we propose an effective data reduction strategy based on Pointwise 𝒱-Information (PVI). To enable a static method, we first use PVI to quantify instance difficulty and remove instances with low difficulty. Experiments show that classifier performance is maintained with only a 0.0001% to 0.76% decline in accuracy when 10–30% of the data is removed. Second, we train the classifiers using a progressive learning strategy on examples sorted by increasing PVI, accelerating convergence and achieving a 0.8% accuracy gain over conventional training. Our findings imply that training a classifier on the chosen optimal subset may improve model performance and increase training efficiency when combined with an efficient data reduction strategy. Furthermore, we have adapted the PVI framework, which was previously limited to English datasets, to a variety of Chinese Natural Language Processing (NLP) tasks and base models, yielding insightful results for faster training and cross-lingual data reduction. Full article

Large, high-quality patient datasets are essential for applications like economic modeling and patient simulation. However, real-world data is often inaccessible or incomplete. Synthetic patient data offers an alternative, and current methods often fail to preserve clinical plausibility, real-world correlations, and logical consistency. This study presents a patient cohort generator designed to produce realistic, statistically valid synthetic datasets. The generator uses predefined probability distributions and Cholesky decomposition to reflect real-world correlations. A dependency matrix handles variable relationships in the right order. Hard limits block unrealistic values, and binary variables are set using percentiles to match expected rates. Validation used two datasets, NHANES (2021–2023) and the Framingham Heart Study, evaluating cohort diversity (general, cardiac, low-dimensional), data sparsity (five correlation scenarios), and model performance (MSE, RMSE, R², SSE, correlation plots). Results demonstrated strong alignment with real-world data in central tendency, dispersion, and correlation structures. Scenario A (empirical correlations) performed best (R² = 86.8–99.6%, lowest SSE and MAE). Scenario B (physician-estimated correlations) also performed well, especially in a low-dimensions population (R² = 80.7%). Scenario E (no correlation) performed worst. Overall, the proposed model provides a scalable, customizable solution for generating synthetic patient cohorts, supporting reliable simulations and research when real-world data is limited. While deep learning approaches have been proposed for this task, they require access to large-scale real datasets and offer limited control over statistical dependencies or clinical logic. Our approach addresses this gap. Full article

This study evaluates the replacement of petroleum-based naphthenic oil with four types of soybean-derived alternatives—virgin soybean oil (SBO), epoxidized SBO (ESBO), expired SBO, and recycled SBO—in styrene–butadiene rubber (SBR) composites. The materials were tested in both staining rubber (SR) and non-staining rubber (NSR) systems to assess processing characteristics, mechanical performance, and environmental durability. Among the alternatives, SBO demonstrated the best overall performance, improving processability and tensile strength by over 10%, while ESBO enhanced ozone resistance by 35% due to its epoxide functionality. Expired and recycled SBOs maintained essential mechanical properties within 90% of virgin SBO values. The full replacement of CH450 with SBO in tire prototypes resulted in burst strength exceeding 1000 kPa and stable appearance after 5000 km of road testing. To validate industrial relevance, the developed green tire was exhibited at the 2025 Taipei International Cycle Show, attracting interest from international buyers and stakeholders for its eco-friendly composition and carbon footprint reduction potential, thereby demonstrating both technical feasibility and commercial viability. Full article