Search Results (1,569)

The rapid growth of digital communication has intensified spam-related threats, including phishing and malware, which employ advanced evasion tactics. Traditional filtering methods struggle to keep pace, driving the need for sophisticated machine learning (ML) solutions. The effectiveness of ML models hinges on selecting high-quality input features, especially in high-dimensional datasets where irrelevant or redundant attributes impair performance and computational efficiency. Guided by principles of symmetry to achieve an optimal balance between model accuracy, complexity, and interpretability, this study proposes an Enhanced Hybrid Quantum-Inspired Firefly and Artificial Bee Colony (EHQ-FABC) algorithm for feature selection in spam detection. EHQ-FABC leverages the Firefly Algorithm’s local exploitation and the Artificial Bee Colony’s global exploration, augmented with quantum-inspired principles to maintain search space diversity and a symmetrical balance between exploration and exploitation. It eliminates redundant attributes while preserving predictive power. For interpretability, Shapley Additive Explanations (SHAPs) are employed to ensure symmetry in explanation, meaning features with equal contributions are assigned equal importance, providing a fair and consistent interpretation of the model’s decisions. Evaluated on the ISCX-URL2016 dataset, EHQ-FABC reduces features by over 76%, retaining only 17 of 72 features, while matching or outperforming filter, wrapper, embedded, and metaheuristic methods. Tested across ML classifiers like CatBoost, XGBoost, Random Forest, Extra Trees, Decision Tree, K-Nearest Neighbors, Logistic Regression, and Multi-Layer Perceptron, EHQ-FABC achieves a peak accuracy of 99.97% with CatBoost and robust results across tree ensembles, neural, and linear models. SHAP analysis highlights features like domain_token_count and NumberOfDotsinURL as key for spam detection, offering actionable insights for practitioners. EHQ-FABC provides a reliable, transparent, and efficient symmetry-aware solution, advancing both accuracy and explainability in spam detection. Full article

Rapid sand filtration is a critical step in the water treatment process, as its effectiveness directly impacts the supply of safe drinking water. However, optimising filtration processes in water treatment plants (WTPs) presents a significant challenge due to the varying operational parameters and conditions. This study applies explainable machine learning to enhance insights into predicting direct filtration operations at the Ålesund WTP in Norway. Three baseline models (Multiple Linear Regression, Support Vector Regression, and K-Nearest Neighbour (KNN)) and three ensemble models (Random Forest (RF), Extra Trees (ET), and XGBoost) were optimised using the GridSearchCV algorithm and implemented on seven filter units to predict their filtered water turbidity. The results indicate that ML models can reliably predict filtered water turbidity in WTPs, with Extra Trees models achieving the highest predictive performance (R² = 0.92). ET, RF, and KNN ranked as the three top-performing models using Alternative Technique for Order of Preference by Similarity to Ideal Solution (A-TOPSIS) ranking for the suite of algorithms used. The feature importance analysis ranked the filter runtime, flow rate, and bed level. SHAP interpretation of the best model provided actionable insights, revealing how operational adjustments during the ripening stage can help mitigate filter breakthroughs. These findings offer valuable guidance for plant operators and highlight the benefits of explainable machine learning in water quality management. Full article

High-resolution CO₂ fossil fuel emission data are critical for developing targeted mitigation policies. As a key approach for estimating spatial distributions of CO₂ emissions, top–down methods typically rely upon spatial proxies to disaggregate administrative-level emission to finer spatial scales. However, conventional linear regression models may fail to capture complex non-linear relationships between proxies and emissions. Furthermore, methods relying on nighttime light data are mostly inadequate in representing emissions for both industrial and rural zones. To address these limitations, this study developed a multiple proxy framework integrating nighttime light, points of interest (POIs), population, road networks, and impervious surface area data. Seven machine learning algorithms—Extra-Trees, Random Forest, XGBoost, CatBoost, Gradient Boosting Decision Trees, LightGBM, and Support Vector Regression—were comprehensively incorporated to estimate high-resolution CO₂ fossil fuel emissions. Comprehensive evaluation revealed that the multiple proxy Extra-Trees model significantly outperformed the single-proxy nighttime light linear regression model at the county scale, achieving R² = 0.96 (RMSE = 0.52 MtCO₂) in cross-validation and R² = 0.92 (RMSE = 0.54 MtCO₂) on the independent test set. Feature importance analysis identified brightness of nighttime light (40.70%) and heavy industrial density (21.11%) as the most critical spatial proxies. The proposed approach also showed strong spatial consistency with the Multi-resolution Emission Inventory for China, exhibiting correlation coefficients of 0.82–0.84. This study demonstrates that integrating local multiple proxy data with machine learning corrects spatial biases inherent in traditional top–down approaches, establishing a transferable framework for high-resolution emissions mapping. Full article

Raptors, being at top of the food chain, serve as important models to study the impact of changing climate, as they are more vulnerable due to their unique ecology. They are vulnerable to extinction, with 52% species declining population and 18% are threatened globally. The effect of climate change on raptors is poorly studied in the Eastern Himalayan region. The present study offers a complete investigation of climate change effects on the raptors in the northeast region of the Eastern Himalayas, employing ensemble species distribution modeling. The future predictions were employed to model the climate change across two socioeconomic pathways (SSP) i.e. SSP245 and SSP585 for the periods 2041–2060 and 2061–2080. Specifically, five algorithms were employed for the ensemble model, viz. boosted regression tree (BRT), generalized linear model (GLM), multivariate adaptive regression splines (MARS), maximum entropy (MaxEnt) and random forest (RF). The study highlights worrying results, as only 10.5% area of the NE region is presently suitable for Falco severus, 11.4% for the critically endangered Gyps tenuirostris, and a mere 6.9% area is presently suitable for the endangered Haliaeetus leucoryphus. The most influential covariates were precipitation of the driest quarter, precipitation of the wettest month, and temperature seasonality. Future projection revealed reduction of 33–41% in suitable habitats for F. severus, G. tenuirostris is expected to lose 53–96% of its suitable habitats, and H. leucoryphus has lost nearly 94–99% of its suitable habitats. Such decline indicates apparent habitat fragmentation, with shrinking habitat patches. Full article

A rapid targeted screening method for 22 compounds, including flavonoids, glycosides, and phenolics, in Dendrobium officinale was developed using UHPLC–MS/MS, demonstrating good linear correlation coefficients, precision, repeatability, and stability. D. officinale from the Guangnan and Maguan regions can be effectively classified into two distinct categories using PCA. In addition, OPLS-DA discriminant analysis enables clear separation between groups, with samples forming well-defined clusters. The 22 chemical components provide valuable origin-related information for D. officinale. The compounds with VIP values of >1 included eriodictyol, vanillic acid, protocatechuic acid, gentisic acid, and naringenin. The difference in naringenin content between D. officinale from the two production areas was minimal. By contrast, eriodictyol and vanillic acid were relatively abundant in D. officinale from Guangnan, while gentisic acid and protocatechuic acid were more prevalent in D. officinale from Maguan. The pathways with higher Kyoto Encyclopedia of Genes and Genomes enrichment were primarily associated with lipid metabolism and atherosclerosis, fluid shear stress and atherosclerosis, and nonalcoholic fatty liver disease. These findings suggest that D. officinale exhibits promising lipid-balancing properties and potential cardiovascular health benefits. Seven machine learning algorithms—Random Forest, XGBoost, Support Vector Machine, k-Nearest Neighbor, Backpropagation Neural Network, Random Tree, and CatBoost—demonstrated superior accuracy and precision in distinguishing D. officinale from the Guangnan and Maguan regions. The key compounds with higher weights—vanillic acid, chrysoeriol, trigonelline, isoquercitrin, gallic acid, 4-hydroxybenzaldehyde, eriodictyol, sweroside, apigenin, and homoeriodictyol—play a crucial role in model construction and the identification of D. officinale from the Guangnan and Maguan regions. The quantification of 22 compounds using UHPLC–MS/MS, combined with PCA, OPLS-DA, and machine learning, enables effective discrimination of D. officinale from these two Yunnan production areas. Full article

Optimising spraying operations in coffee cultivation can enhance both application efficiency and effectiveness. However, no studies have specifically assessed droplet deposition on leaves adjacent to the spray application band—fraction of droplet deposition referred to as ‘transfer’ in this study. Therefore, this study aimed to quantify droplet deposition and transfer resulting from different application rates and nozzle types in coffee trees. The experiment was conducted in a factorial design including three application rates (200, 400, and 600 L ha⁻¹) and two nozzle types (hollow cone and flat fan), with four replicates. Deposition was quantified at multiple positions: two application sides (left and right), three sections of the plant (upper, middle, and lower), and two branch positions (inner and outer). Thus, all measurements across sides, plant sections, and branch positions were nested, resulting in correlated data that were analysed using linear mixed-effects models (lme4 package), with parameters estimated using the restricted maximum likelihood method. The flat fan nozzle achieved the highest reference deposition, particularly on outer canopy thirds, while spray transfer (~29% of total deposition) was mainly driven by operational factors. Hollow cone nozzles at 200 L ha⁻¹ minimized transfer while maintaining adequate deposition. Optimizing applications requires maximizing reference deposition and minimizing transfer, which can be achieved through operational adjustments, airflow management, and complementary strategies such as adjuvants, electrostatic spraying, or tunnel sprayers. Full article

Wave-powered desalination systems integrate reverse osmosis (RO) with renewable ocean energy, providing a sustainable and environmentally responsible approach to freshwater production. This study aims to investigate wave-powered RO desalination using supervised and deep machine learning (ML) models to predict the effects of variable feed flow on permeate recovery and salt rejection under dynamic hydrodynamic conditions. Multiple ML models, including Gaussian process regression (GPR), support vector machines (SVMs), multi-layer perceptron (MLP), linear regression (LR), and decision trees (DTs) were systematically assessed for the prediction of permeate recovery and salt rejection (%) using three distinct input configurations: limited physicochemical features (M1), flow- and salinity-related parameters (M2), and a comprehensive variable set incorporating temperature (M3). GPR achieved near-perfect predictive accuracy R² values (~1.00) with minimal errors for permeate recovery and salt rejection, attributed to its flexible kernel and probabilistic design. MLP and SVM also performed well, though they showed greater sensitivity to feature complexity. In contrast, DT models exhibited limited generalization and higher error rates, particularly when key features were excluded. Sensitivity analyses revealed that feed pressure (FP) and brine salinity (BS) were dominant positive influencers of permeate recovery and salt rejection. In contrast, brine flow (BF) and permeate salinity (PS) had negative impacts. Full article

Urban traffic noise has become an increasingly significant environmental and public health issue, with many cities—particularly those experiencing rapid urban growth, such as Kuwait—recording levels that often exceed recommended limits. In this study, we present a detailed, data-driven approach for assessing and predicting equivalent continuous noise levels (LAeq) in residential neighborhoods. The analysis draws on measurements taken at 12 carefully chosen sites covering different road types and urban settings, resulting in 21,720 matched observations. A range of predictors was considered, including road classification, traffic composition, meteorological variables, spatial context, and time of day. Four predictive models—Linear Regression, Support Vector Machine (SVM), Gaussian Process Regression, and Bagged Trees—were evaluated through 5-fold cross-validation. Among these, the Bagged Trees model achieved the strongest performance (R² = 0.91, RMSE = 2.13 dB(A)). To better understand how the model made its predictions, we used SHAP (SHapley Additive Explanations) analysis, which showed that road classification, location, heavy vehicle volume, and time of day had the greatest influence on noise levels. The results identify the main determinants of traffic noise in Kuwait’s urban areas and emphasize the role of targeted design and planning in its mitigation. Full article

This study presents the deployment of Machine Learning (ML) models on low-cost edge devices (ESP32) for real-time water quality prediction in tilapia aquaculture. A compact monitoring and control system was developed with low-cost sensors to capture key environmental parameters under field conditions in Northern Thailand. Three ML models—Multiple Linear Regression (MLR), Decision Tree Regression (DTR), and Random Forest Regression (RFR)—were evaluated. RFR achieved the highest accuracy (R² > 0.80), while MLR, with moderate performance (R² ≈ 0.65–0.72), was identified as the most practical choice for ESP32 deployment due to its computational efficiency and offline operability. The system integrates sensing, prediction, and actuation, enabling autonomous regulation of dissolved oxygen and pH without constant cloud connectivity. Field validation demonstrated the system’s ability to maintain DO within biologically safe ranges and stabilize pH within an hour, supporting fish health and reducing production risks. These findings underline the potential of Edge AIoT as a scalable solution for small-scale aquaculture in resource-limited contexts. Future work will expand seasonal data coverage, explore federated learning approaches, and include economic assessments to ensure long-term robustness and sustainability. Full article

Origin–destination (OD) passenger flow is a critical variable for metro system planning and operation. While numerous studies have investigated the influence of the built environment on passenger flow, most have focused on ingress or egress flows at metro stations. The impact of the built environment on OD flow dynamics, particularly the differences between origin-side and destination-side effects, remains poorly understood. This study proposes a novel method for exploring the non-linear effects of station-area built environments on OD flow in large-scale metro networks. First, hourly OD flows and station-area built environment features were extracted from multi-source data. Next, an analytical framework was developed to model the built environment–OD flow relationship using a gradient boosting decision tree model. Finally, the contributions of built environment variables and their non-linear effects on OD flows were systematically investigated. The proposed method was implemented on the Suzhou metro network in China. Test results show that most built environment variables exhibit time-varying, non-linear correlations with OD flows. Even the same variable demonstrates notable differences in its effect between the origin and destination sides. The findings of this study provide valuable guidance for metro planning and station-area urban development. Full article

Slope stability prediction is a critical task in geotechnical engineering, but machine learning (ML) models require large datasets, which are often costly and time-consuming to obtain. This study proposes a domain-driven teacher–student framework to overcome data limitations for predicting the dry factor of safety (FS dry). The teacher model, XGBoost, was trained on the original dataset to capture nonlinear relationships among key site-specific features (unit weight, cohesion, friction angle) and assign pseudo-labels to synthetic samples generated via domain-driven simulations. Six student models, random forest (RF), decision tree (DT), shallow artificial neural network (SNN), linear regression (LR), support vector regression (SVR), and K-nearest neighbors (KNN), were trained on the augmented dataset to approximate the teacher’s predictions. Models were evaluated using a train–test split and five-fold cross-validation. RF achieved the highest predictive accuracy, with an R² of up to 0.9663 and low error metrics (MAE = 0.0233, RMSE = 0.0531), outperforming other student models. Integrating domain knowledge and synthetic data improved prediction reliability despite limited experimental datasets. The framework provides a robust and interpretable tool for slope stability assessment, supporting infrastructure safety in regions with sparse geotechnical data. Future work will expand the dataset with additional field and laboratory tests to further improve model performance. Full article

As global climate change intensifies the frequency of extreme weather events, urban flood control and disaster reduction efforts face unprecedented challenges. With the limitations of traditional, top-down emergency management becoming increasingly apparent, many countries are actively incorporating community-based participation into flood risk governance. While research in this area is expanding, the specific impact of urban flood inundation severity on residents’ participation in relief efforts remains significantly underexplored. To address this research gap, this study employs the Community Capitals Framework (CCF) and a Gradient Boosting Decision Tree (GBDT) model to empirically analyze 1322 survey responses from Zhengzhou, China, exploring the non-linear relationship between flood severity and public participation. Our findings are threefold: (1) As the most direct source of residents’ risk perception, flood inundation severity has a significant association with their participation level. (2) This relationship is distinctly non-linear. For instance, inundation severity within a 200 m radius of a resident’s home shows a predominantly negative relation with participation level, with the negative effect lessening at extreme levels of inundation. The distance from inundated areas, conversely, exhibits an “S-shaped” curve. (3) Flood severity exhibits a significant reinforcement interaction with both communication technology levels and government organizational mobilization. This indicates that, during public crises like flash floods, robust information channels and effective organizational support are positively related to residents’ transition from passive to active participation. This study reveals the complex, non-linear associations between flood severity and civic engagement, providing theoretical support and practical insights for optimizing disaster policies and enhancing community resilience within the broader context of urban land management and sustainable development. Full article

Urban soundscapes are increasingly recognized as fundamental for both ecological integrity and human well-being, yet the complex interplay between the vegetation structure, seasonal dynamics, and microclimatic factors in shaping these soundscapes remains poorly understood. This study tests the hypothesis that vegetation structure and seasonally driven biological activity mediate the balance and the quality of the urban acoustic environment. We investigated seasonal and spatial variations in five acoustic indices (NDSI, ACI, AEI, ADI, and BI) within a historical urban garden in Castelfranco Veneto, Italy. Using linear mixed-effects models, we analyzed the effects of season, microclimatic variables, and vegetation characteristics on soundscape composition. Non-parametric tests were used to assess spatial differences in vegetation metrics. Results revealed strong seasonal patterns, with spring showing increased NDSI (+0.17), ADI (+0.22), and BI (+1.15) values relative to winter, likely reflecting bird breeding phenology and enhanced biological productivity. Among microclimatic predictors, temperature (p < 0.001), humidity (p = 0.014), and solar radiation (p = 0.002) showed significant relationships with acoustic indices, confirming their influence on both animal behaviour and sound propagation. Spatial analyses showed significant differences in acoustic patterns across points (Kruskal–Wallis p < 0.01), with vegetation metrics such as tree density and evergreen proportion correlating with elevated biophonic activity. Although the canopy height model did not emerge as a significant predictor in the models, the observed spatial heterogeneity supports the role of vegetation in shaping urban sound environments. By integrating ecoacoustic indices, LiDAR-derived vegetation data, and microclimatic parameters, this study offers novel insights into how vegetational components should be considered to manage urban green areas to support biodiversity and foster acoustically restorative environments, advancing the evidence base for sound-informed urban planning. Full article

Accurate identification of plastic polymers is essential for effective recycling, quality control, and environmental monitoring. This study assesses how spectral derivative preprocessing affects the classification of six common plastic polymers: Polyethylene Terephthalate (PET), Polyvinyl Chloride (PVC), Polypropylene (PP), Polystyrene (PS), and both High- and Low-Density Polyethylene (HDPE and LDPE), based on Fourier Transform Infrared (FTIR) spectroscopy data acquired at a resolution of 8 ${cm}^{-1}$. Using Savitzky–Golay derivatives (orders 0, 1, and 2), five machine learning algorithms, namely Multilayer Perceptron (MLP), Extremely Randomized Trees (ET), Linear Discriminant Analysis (LDA), Support Vector Classifier (SVC), and Random Forest (RF), were tested within a strict framework involving stratified repeated cross-validation and a final hold-out test set to evaluate generalization. The first spectral derivative notably improved the model performance, especially for MLP and SVC, and increased the stability of the ET, LDA, and RF classifiers. The combination of the first derivative with the ET model provided the best results, achieving a mean F1-score of 0.99995 (±0.00033) in cross-validation and perfect classification (1.0 in Accuracy, F1-score, Cohen’s Kappa, and Matthews Correlation Coefficient) on the independent test set. LDA also performed very well, underscoring the near-linear separability of spectral data after derivative transformation. These results demonstrate the value of derivative-based preprocessing and confirm a robust method for creating high-precision, interpretable, and transferable machine learning models for automated plastic polymer identification. Full article

Traditional machine learning approaches for 5G network management relieve data from operational networks, which are often noisy and confounded, making it difficult to identify key influencing factors. This research addresses the critical gap between correlation-based prediction and interpretable, data-driven explanation. To this end, a software-defined standalone 5G architecture was developed using srsRAN and Open5GS to support multi-user scenarios. A multi-user environment was then simulated with GNU Radio, from which the initial dataset was collected. This dataset was further generated using a Conditional Tabular Generative Adversarial Network (CTGAN) to improve diversity and balance. Several machine learning models, including Linear Regression, Decision Tree, Random Forest, Gradient Boosting, and XGBoost, were trained and evaluated for predicting network performance. Among them, XGBoost achieved the best results, with an R² score of 0.998. To interpret the model, we conducted a SHAP (SHapley Additive exPlanations) analysis, which revealed that the download-to-upload bitrate ratio (dl_ul_ratio) and upload bitrate (brate_ul) were the most influential features. By leveraging a controlled experimental 5G environment, this study demonstrates how machine learning can move beyond predictive accuracy to uncover the fundamental principles governing 5G system performance, providing a robust foundation for future network optimization. Full article