Search Results (785)

Apple pomace represents a sustainable source of phenolic compounds with significant antioxidant potential. This study investigated the recovery of water-extractable bioactive constituents from apple pomace using aqueous ultrasound-assisted extraction (UAE) combined with biochemical profiling and machine learning-based modelling. Extraction conditions were varied according to time (10–30 min), temperature (25–75 °C), and solvent-to-solid ratio (10–20 mL/g). The obtained extracts were evaluated for total phenolic content (TP), total flavonoid content (TF), antioxidant capacity (DPPH, ABTS, CUPRAC, FRAP, metal chelating, and phosphomolybdenum assays), and enzyme inhibitory activities against acetylcholinesterase, butyrylcholinesterase, tyrosinase, α-amylase, and α-glucosidase. TP and TF ranged from 4.70 to 9.97 mg GAE/g and 0.21–0.79 mg RE/g, respectively, while antioxidant assays demonstrated substantial variation depending on extraction conditions. Strong correlations (r = 0.827–0.945, p < 0.001) were observed between phenolic content and antioxidant activity. Artificial neural networks, random forests, support vector machines, and a hybrid ensemble model were applied to predict extraction outcomes, with predictive performance varying substantially among biochemical responses and modelling approaches. Multi-objective optimization using the NSGA-II formulation identified a representative Pareto compromise at approximately 15.1 min, 25.7 °C, and a solvent-to-solid ratio of 13.7 mL/g, balancing desirable biochemical responses with processing requirements. Full article

Many existing deep-learning models for brain-tumor diagnosis report very high accuracy, yet they often suffer from overfitting, limited generalizability, and poor robustness, which reduces their clinical reliability. First, a large and diverse dataset was assembled from multiple publicly available sources to reduce data bias. Next, a series of regularization and calibration techniques were applied, including label smoothing, mixup augmentation ($\alpha =0.4$), exponential moving average (EMA) of model weights to prevent overfitting, stabilize training, and temperature scaling to calibrate predicted probabilities. Since brain-tumor diagnosis directly affects critical clinical decisions, both model accuracy and reliability of its predictions are essential. For that, six diverse, powerful pretrained backbones (ConvNeXt, EfficientNetV2, DenseNet201, Swin, ViT, and DINOv3) were used as frozen feature extractors under pure transfer learning. On top of these frozen features, lightweight classification heads were trained using concatenation, attention, and gated fusion. Then, the predicted-class probabilities from the selected heads were combined and passed to a logistic-regression meta-learner to produce the final prediction, with the aim of improving accuracy and reliability. This architecture is computationally efficient because only lightweight classifiers and a logistic meta-learner are trained, while the heavy backbones remain frozen. The best-performing variant on TEST2, attn_fp_ens , was further evaluated using synchronized 10-fold cross-validation on the TRAIN set. The best-performing TEST variant, the gated-fusion ensemble with fingerprint features (gated_fp_ens), achieved 99.33% test accuracy and 99.28% macro-F1. Under domain shift on the external hospital dataset (TEST2), the attention-fusion ensemble with fingerprint features (attn_fp_ens) showed the strongest performance among the evaluated variants, achieving 95.16% accuracy and 94.11% macro-F1. The TEST2 dataset was collected from a local hospital for external evaluation. Full article

Magnetocardiography (MCG) is highly susceptible to noise during acquisition, which significantly limits its clinical utility for the assessment of Coronary Artery Disease (CAD). Traditional model-driven and data-driven denoising methods often suffer from manual parameter tuning, inadequate feature preservation, and limited adaptability, resulting in suboptimal performance in noisy environments. To address these challenges, this paper introduces a novel hybrid model-driven and data-driven framework called LBTO-VMD-DLSTM. The framework utilizes a logistic-map-enhanced Bermuda Triangle Optimization (LBTO) algorithm to adaptively optimize the parameters of Variational Mode Decomposition (VMD), enabling effective separation of signal and noise. Furthermore, the Dual-head attention mechanism embedded in a Long Short-Term Memory (DLSTM) network captures long-range temporal dependencies, while the network is trained using a supervised learning method to learn the mapping from noisy Intrinsic Mode Functions (IMFs) to clean signals, ultimately achieving denoising without requiring manual component selection. To verify the superiority of LBTO-VMD-DLSTM, its performance was compared with traditional algorithms such as Empirical Mode Decomposition (EMD), Denoising Autoencoder (DAE), and Complete Ensemble Empirical Mode Decomposition with Adaptive Noise combined with Wavelet Transform (CEEMDAN-WT). The experimental results show that the proposed denoising method achieved a maximum Signal-to-Noise Ratio (SNR) of 21.75 dB and a Cosine Similarity (CosSim) of 0.99. In tests conducted on MCG signals with added noise at different SNR levels and different types of baseline drift, the LBTO-VMD-DLSTM method produced the highest SNRs (20.91 dB and 23.5 dB) and the highest CosSim values (0.99 and 0.98). This indicates that the proposed method outperforms the comparison algorithms, effectively eliminating complex noise in MCG signals while preserving waveform features critical for subsequent clinical interpretation. Full article

The last few years have witnessed an exponential rise in the demand of precision agriculture that needs a robust agroclimatic mapping. The nonlinearity, spatial, temporal and region-specific climate interactions which governs the production makes the crop–climate mapping a challenging task. Centered on areca nut yield, in the present study, a novel spatiotemporal integrated hybrid RF attention-BiLSTM framework is developed to analyze the crop and climate associations in Central Karnataka. By integrating a comprehensive data preprocessing strategy, including missing-value estimation, outlier treatment, scale normalization, data stratification, and SMOTE–ENN resampling, the developed model captured the region-specific climate sensitive production pattern. The hybrid model achieved an R² value of 0.914, RMSE of 0.39, and correlation coefficient of 0.956, outperforming conventional models. The repeated-run, statistical, and ablation analyses further demonstrated the model consistency and the effectiveness of integrated hybrid framework. Further, a genotype analysis is performed to analyze resilience and stability of different areca nut varieties. The overall statistical results confirm robustness of the proposed framework, where the ability to amalgamate data-level optimization, attention-driven temporal learning, and ensemble modeling enables it robust and hence well-suited for complex, real-world time-series agroclimatic mapping and prediction tasks for areca nut precision agriculture decisions. Full article

Android malware continues to evolve in complexity, creating challenges for detection systems that must distinguish malicious applications from increasingly heterogeneous benign applications. Although machine learning provides effective mechanisms for learning malware characteristics, the high dimensionality of Android malware datasets can introduce redundant and weakly informative features and increase computational requirements. To address this problem, this paper proposes a Hybrid Feature Selection and Soft-Voting Ensemble (HFS-SVE) framework that integrates complementary feature-selection and ensemble-learning strategies. The proposed framework sequentially applies Random Forest (RF) feature importance, Chi-square-based SelectKBest, correlation filtering, and L1 regularisation, reducing the original 489-feature representation to 13 selected features. These features are subsequently classified using RF, XGBoost, and LightGBM, whose probability outputs are combined through soft voting. Experimental results on the KronoDroid dataset demonstrate that the proposed HFS-SVE achieves 99.41% accuracy, 99.52% precision, 99.30% recall, 99.40% F1-score, and 99.41% ROC-AUC. The proposed framework also records the lowest measured detection time among the evaluated feature-selection strategies. Cross-dataset evaluation on Malgenome, TUANDROMD, and Drebin achieves accuracy above 98% on each dataset. The findings demonstrate that the proposed HFS-SVE can combine substantial feature-space reduction with strong Android malware detection performance, while the cross-dataset results highlight the importance of dataset variation and feature provenance when assessing generalisation. Full article

Reliable assessment of weather accessibility is essential for offshore operations, particularly in areas such as the southern Baltic Sea, characterized by fetch-limited and highly variable wave conditions. This study combines a methodological review of weather-window approaches with operational validation of deterministic threshold methods. The review identified seven methodological groups: deterministic threshold, probabilistic, ensemble, hindcast-based, stochastic, machine learning, and hybrid approaches. Probabilistic, ensemble, and hybrid approaches offer greater potential for short-term forecasting, whereas deterministic methods were considered suitable for long-term strategic planning. This approach was therefore selected for validation against operational survey records. The framework uses 82 years of hourly ERA5 reanalysis data, significant wave height thresholds (Hs ≤ 0.5, 1.0, 1.5, and 2.0 m), and a 24 h weather-window criterion. Model-derived monthly downtime was compared with observed weather standby hours documented in project records for 40 monthly observations from three survey vessels operating in the Polish Exclusive Economic Zone between 2022 and 2026. The model overestimated monthly downtime, with a mean bias error (MBE) of +73.7 h, MAE = 108.4 h, and RMSE = 135.9 h. The 95% limits of agreement ranged from −152.9 to +300.4 h, with r = 0.764 and R² = 0.584. The results indicate that the framework is more suitable for long-term strategic assessment of weather accessibility than for precise prediction of downtime during individual survey campaigns. Full article

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder for which accurate and accessible screening approaches remain an important research objective. This study proposes a hybrid meta-classifier framework for handwriting-based AD classification using the DARWIN (Diagnosis AlzheimeR WIth haNdwriting) dataset, which comprises 174 participants, including 89 individuals with AD and 85 healthy controls, and contains 450 handwriting-related features derived from 25 tasks. The proposed framework combines a one-dimensional convolutional neural network (1D-CNN) for automated feature learning with a heterogeneous ensemble of XGBoost, support vector machine (SVM), and random forest classifiers. The predictions of the base classifiers are subsequently integrated by an AdaBoost-based meta-classifier to generate the final classification decision. The framework achieved a pooled cross-validation accuracy of 98.28%, with a mean fold-level accuracy of 97.73 ± 1.27%, a precision of 96.90 ± 2.84%, a recall of 98.89 ± 2.49%, an F1-score of 97.84 ± 1.21%, and an AUC of 0.944 ± 0.023 across the five outer folds. In addition, evaluation on an independent handwriting-signature cohort achieved an accuracy of 91.76%, with a sensitivity of 90.74% and a specificity of 93.55%. These results indicate that the proposed framework has strong discriminative capability across the evaluated datasets. Nevertheless, further validation on larger and more diverse independent cohorts is required before conclusions regarding clinical applicability can be drawn. Future work will investigate multimodal integration and the application of explainable artificial intelligence techniques to improve the interpretability of the proposed framework. Full article

Accurate forecasting of electrical load at the substation level is essential for detecting grid losses that pose significant financial and safety challenges. This study investigates whether the standard correction method currently used by grid operators can be improved through the application of advanced forecasting models to granular, location-specific load data. Such improvements are particularly valuable for identifying abnormal consumption patterns indicative of grid losses due to electricity theft, defective meters, or cable damage. This paper evaluates a broad range of statistical and machine learning models—including ARIMAX, SARIMAX, Random Forests, Gradient Boosting Machines, Neural Networks, and Support Vector Regression—based on unique quarter-hourly datasets from several Dutch substations. Two hybrid approaches are proposed, combining the best-performing individual models through a stacked ensemble method and a simpler averaging strategy. The results show that incorporating lagged and additional exogenous variables, along with the application of various advanced models, significantly improves forecasting accuracy compared to the standard correction method, with the best hybrid model reducing MAE and RMSE by approximately 64.7% and 61.6%, respectively, relative to the current operational benchmark. This study demonstrates that substation-level, data-driven forecasting can strengthen the signals used to detect grid losses, offering practical implications for grid operators and policymakers. Full article

Student engagement within Learning Management Systems (LMSs) provides valuable behavioural data for understanding and predicting learning outcomes. However, predicting students’ forum consumption behaviour remains challenging because LMS datasets may contain redundant engagement indicators that increase model complexity. This study proposes a hybrid predictive modelling framework that integrates Ant Colony Optimisation (ACO) with three ensemble regression algorithms—Random Forest (RF), Gradient Boosting (GB), and Stacking—to predict forum consumption behaviour using LMS-derived engagement indicators. Guided by Educational Data Mining (EDM) and Social Learning Theory (SLT), behavioural, cognitive, and social engagement dimensions were operationalised using LMS indicators, with Freq_Forum_Consume serving as the target variable. ACO was employed as a wrapper-based feature-selection technique to identify informative predictors before model training. The performance of the ACO–ensemble models was compared with corresponding baseline models using the coefficient of determination (R²), Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE). The results show that ACO reduced the predictor space from nine to six variables for GB and to seven variables for both Stacking and RF, while maintaining or improving predictive performance. ACO-GB achieved the strongest overall performance (R² = 0.8332, MAE = 55.3173, RMSE = 71.4150). Consistent results across multiple ACO parameter configurations further demonstrated parameter consistency within the tested search settings. The selected predictors represented behavioural, cognitive, and social engagement dimensions, highlighting their complementary contribution to predicting forum consumption behaviour. The proposed framework provides a more parsimonious and interpretable approach to LMS-based learning analytics while retaining predictive performance. Full article

This study examines how accounting professionals’ self-reported forensic-auditing capabilities relate to their attitudinal readiness to recognize suspicious anti-money-laundering (AML) indicators, using a transparent hybrid design that combines classical psychometrics, structural equation modeling (SEM), and leakage-controlled explainable machine learning (ML). A cross-sectional survey collected 700 valid responses from public accountants in Lima, Peru, with a 30-item, five-point Likert instrument (19 forensic-auditing items in three dimensions and 11 AML-recognition items). Reliability was high (Cronbach’s α = 0.935; McDonald’s ω = 0.935), but average variance extracted was below 0.50 in every block (0.329–0.449), and Fornell–Larcker testing showed that skills-and-knowledge and AML recognition were not discriminantly distinct (r = 0.673 > √AVE = 0.651/0.649). Responses showed a pronounced ceiling (51% of answers were the maximum), and 86 respondents (12.3%) answered all 30 items identically; removing them lowered the forensic-auditing–AML association from r = 0.731 to 0.630 and explained variance from 53.8% to 40.3%. Skills-and-knowledge remained the strongest predictor in SEM and HC3-robust regression (β = 0.478 and 0.425). Under a leakage-free protocol, ensemble models reached ROC-AUC ≈ 0.86 on held-out data, but threshold tuning did not improve F1 test, and item-level attributions were unstable (Spearman ρ = 0.28). Forensic-auditing capabilities are positively associated with declared AML-recognition readiness, driven by applied skills and knowledge; the evidence is attitudinal and correlational, and should not be read as real detection capability. Because professional experience, seniority, sector, and prior AML training were not measured, the reported associations may be partly confounded by unobserved professional background, and the dominance of skills-and-knowledge is therefore advanced as tentative, pending resolution of the skills-and-knowledge/AML-recognition discriminant-validity overlap. A second, procedural contribution is that the study reports the data-quality screening, the failed validity tests, and the explanation-stability diagnostics that survey-based forensic-accounting research rarely makes visible. Full article

Landslide susceptibility assessment is essential for maintaining resilient mountain transport corridors where slope failures can disrupt mobility, freight movement, and economic activity. This study evaluated standalone and hybrid machine-learning models for the Kamchik Pass corridor of Uzbekistan, which carries the A-373 Tashkent–Osh highway. Thirteen topographic, hydrological, geological, climatic, vegetation, and land-cover factors were integrated with a landslide inventory. K-nearest neighbours (KNN), Random Forest (RF), XGBoost, and artificial neural network (ANN) models were compared to three RF-based hybrids: RF + KNN, RF + XGBoost, and RF + ANN. Performance was assessed using confusion-matrix metrics and receiver operating characteristic area under the curve (ROC–AUC), together with variable-importance and class-area analyses. All predictors were retained because variance inflation factors remained below 3.3. Slope and elevation were the most consistent predictors across the standalone models. RF + KNN achieved the best performance, with an accuracy of 0.8429, kappa of 0.6857, sensitivity of 0.8857, specificity of 0.8000, and AUC of 0.88. Its map classified 16.94 km², approximately 10.8% of the study area, as high or very high susceptibility. Compared to standalone RF, RF + KNN increased accuracy by 2.86 percentage points and AUC by 0.03, while sensitivity decreased from 0.9429 to 0.8857 and specificity increased from 0.6857 to 0.8000. Because these differences were obtained from a small point-level hold-out set without spatially independent validation, they are interpreted as descriptive rather than evidence of universal model superiority. The maps provide a first-pass susceptibility screening layer for subsequent field verification and asset-exposure analysis; they do not constitute an implemented infrastructure-risk assessment. Full article

Driving style, reflected in recurrent patterns of acceleration, braking, speed selection, following distance, gear use, and lane-changing behavior, plays an important role in road safety and is also associated with fuel consumption, emissions, passenger comfort, and vehicle wear. This paper presents a structured review of recent research on driving style analysis, with particular emphasis on its relationship with road-safety outcomes and risk indicators. Following a PRISMA-oriented methodology, studies published between 2015 and 2025 were identified, screened, and synthesized to examine how driving styles are defined, detected, classified, and evaluated. The review shows a clear shift toward data-driven approaches, including feature-based machine learning and representation-learning methods using support vector machines, ensemble models, convolutional neural networks, recurrent neural networks, and hybrid deep learning architectures. Common data sources include smartphone inertial and GNSS signals, CAN/OBD vehicle data, telematics platforms, naturalistic driving datasets, and camera-based perception systems. Safety impact is most often assessed through crashes, near-miss events, traffic conflicts, time-to-collision measures, harsh maneuvers, and composite risk scores. Across the reviewed literature, aggressive and unstable driving patterns are generally associated with reduced safety margins and increased risk, although comparability remains limited by inconsistent label definitions, heterogeneous datasets, indirect safety proxies, and varied validation protocols. The paper also proposes a reproducible database architecture linking drivers, trips, driving events, and safety events to support transparent analysis, benchmark development, and future implementation in fleet monitoring, driver feedback, and connected vehicle applications. Full article

With the escalating frequency of sophisticated zero-day attacks, overcoming the critical limitations of signature-based Intrusion Detection Systems (IDSs) has become paramount. This study proposes a hybrid multi-layered intrusion detection framework combining traditional machine learning, Deep Neural Architectures (DenseNN), and ensemble methods to evaluate zero-day resilience within cloud-level backend connectivity interfacing EV and V2X management ecosystems. Using the comprehensive CSE-CIC-IDS2018 benchmark as a surrogate environment, a code-executed Leave-One-Attack-Out (LOAO) cross-validation protocol across 13 distinct attack families was implemented to assess unseen-attack-family generalization within the benchmark to unseen threats. Furthermore, Explainable Artificial Intelligence (XAI) auditing, utilizing SHapley Additive exPlanations (SHAP) and Integrated Gradients, was integrated to inspect decision boundaries and resolve feature-attribution failure modes. Critically, the audit identified an artifact-driven data leakage caused by the Timestamp and identifier features, demonstrating that models learned temporal schedules rather than behavioral network signatures. Re-executing all experiments post-leakage removal quantified performance drops across all classifiers (e.g., Gaussian NB dropping by up to 20.88 percentage points in accuracy (at the 60% training ratio; 18.30 points at the 80% ratio)). Under standard binary classification metrics, tree ensembles (Random Forest and Extra Trees) achieved high in-distribution detection (F1 > 0.95) with rapid inference latency (≈0.05–−0.07 ms/sample). However, the rigorous LOAO evaluation revealed a substantial generalization penalty on truly unseen zero-day families (e.g., SQL Injection and Infiltration), where simpler linear models demonstrated broader generalization robustness (mean LOAO F1 = 0.397) compared with complex tree-ensemble models. By rectifying dataset leakage and benchmarking deployment trade-offs (training runtime, throughput, and memory footprint), this study delivers actionable, transparent guidelines for deployment-oriented IDS evaluation in dynamic network infrastructures. Full article

The use of Artificial Intelligence (AI) and Machine Learning (ML) in cybersecurity, especially for creating Intrusion Detection Systems (IDSs), has become increasingly important. These systems are essential for detecting malicious behaviour, identifying network issues, and stopping cyberattacks in real time. Despite extensive research on various ML and Deep Learning (DL) models for IDS, the current literature remains incomplete. It has many different datasets, methods, and evaluation standards. As cyber threats become more advanced, it is crucial to conduct a thorough analysis of ML techniques for intrusion detection. The goal of this Systematic Literature Review (SLR) is to provide a full picture of the most recent academic articles on ML-based IDS. The study addresses important research questions about the most widely used algorithms, the types of attacks and network environments covered, the methodological problems that remain unsolved, and the new trends that should shape future research. Following the PRISMA framework, we conducted a systematic review of peer-reviewed articles published between January 2022 and May 2025. We searched IEEE Xplore, ACM Digital Library, and SpringerLink, yielding 22,558 initial records. After carefully applying strict inclusion criteria, 125 papers were selected for the final analysis. We created a standardised data extraction form (i.e., using MS Excel) to gather bibliographic details, research emphasis, methodological strategies, datasets, evaluation criteria, and recognised constraints. We employed thematic analysis to develop a clear taxonomy. We identified five main research themes in our analysis: (1) ensemble and hybrid learning pipelines focused on performance optimisation (30 papers), (2) context-specific IDS designs for Internet of Things (IoT), cloud, and Software-Defined Networking (SDN) environments (34 papers), (3) data-centric engineering that deals with class imbalance and feature selection (20 papers), (4) deep neural architectures for representation learning (31 papers), and (5) trustworthiness concerns like adversarial robustness, zero-day detection, and Explainable AI (XAI) (10 papers). Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM), and Random Forests are the most commonly used algorithms, often combined. Nonetheless, significant deficiencies remain: about 2% of papers incorporate XAI, only 4% focus on adversarial robustness, and none validate their models in real-world production settings. Denial-of-Service (DoS) and Distributed DoS (DDoS) attacks are the most common types in the literature, whereas Web attacks, ransomware, and advanced persistent threats remain poorly studied. The number of publications grows at an average of 30.2% annually, but the field still relies on legacy benchmark datasets rather than operational validation. Full article

Accurate vibration prediction is critical for condition monitoring and predictive maintenance of hydropower units, yet it remains challenging due to strong noise interference, severe non-stationarity, and multi-scale coupling characteristics of raw vibration signals. This paper proposes a novel hybrid prediction framework that integrates cooperative denoising, multi-scale signal decomposition, and deep learning-based sequential modeling to achieve high-precision long-term vibration forecasting. First, a two-stage cooperative denoising stategy combining wavelet threshold denoising (WTD) and singular spectrum analysis (SSA) is designed to suppress high-frequency noise while effectively preserving the global trend and critical transient features. Then the denoised signal is decomposed into a set of physically interpretable intrinsic mode functions (IMFs) and a residual component via complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), which alleviates mode mixing and improves decomposition completeness. Subsequently, each IMF component is independently predicted using a gated recurrent unit (GRU) network optimized by the Adam algorithm with adaptive learning rate decay, enabling efficient capture of nonlinear temporal dependencies. The proposed framework is validated using 3.5-year real-world vibration data from an lower guide bearing of a pumped-storage hydropower unit. Experimental results demonstrate that the model achieves MAE = 0.3831, RMSE = 0.6964, MAPE = 0.2832%, and $R^2=0.9911$, compared to 0.6666 for the conventional CEEMDAN-GRU model, a 32.45 percentage point increase and a 54% reduction in unexplained variance. Ablation studies and comparative analyses verify the superiority and statistical significance of the cooperative denoising mechanism and the overall hybrid architecture. This work provides a reliable, interpretable, and deployable tool for the condition monitoring and predictive maintenance of hydropower units, supporting proactive operation and reducing unplanned downtime in clean energy systems. Full article