Search Results (114)

Lithofacies, as an integrated representation of key reservoir attributes including mineral composition and organic matter enrichment, provides crucial geological and engineering guidance for identifying “dual sweet spots” and designing fracturing strategies in deep shale gas reservoirs. However, reliable lithofacies characterization remains particularly challenging owing to significant reservoir heterogeneity, scarce core data, and imbalanced facies distribution. Conventional manual log interpretation tends to be cost prohibitive and inaccurate, while existing intelligent algorithms suffer from inadequate robustness and suboptimal efficiency, failing to meet demands for both precision and practicality in such complex reservoirs. To address these limitations, this study developed a super-integrated lithofacies identification model termed SRLCL, leveraging well-logging data and lithofacies classifications. The proposed framework synergistically combines multiple modeling advantages while maintaining a balance between data characteristics and optimization effectiveness. Specifically, SRLCL incorporates three key components: Newton-Weighted Oversampling (NWO) to mitigate data scarcity and class imbalance, the Polar Light Optimizer (PLO) to accelerate convergence and enhance optimization performance, and a Stacking ensemble architecture that integrates five heterogeneous algorithms—Support Vector Machine (SVM), Random Forest (RF), Light Gradient Boosting Machine (LightGBM), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM)—to overcome the representational limitations of single-model or homogeneous ensemble approaches. Experimental results indicated that the NWO-PLO-SRLCL model achieved an overall accuracy of 93% in lithofacies identification, exceeding conventional methods by more than 6% while demonstrating remarkable generalization capability and stability. Furthermore, production simulations of fractured horizontal wells based on the lithofacies-controlled geological model showed only a 6.18% deviation from actual cumulative gas production, underscoring how accurate lithofacies identification facilitates development strategy optimization and provides a reliable foundation for efficient deep shale gas development. Full article

With the continuous growth of user reviews, identifying underlying sentiment across multi-source texts efficiently and accurately has become a significant challenge in NLP. Traditional single models in cross-domain sentiment analysis often exhibit insufficient stability, limited generalization capabilities, and sensitivity to class imbalance. Existing ensemble methods predominantly rely on static weighting or voting strategies among homogeneous models, failing to fully leverage the complementary advantages between models. To address these issues, this study proposes a heterogeneous ensemble sentiment classification model integrating multi-view features and dynamic weighting. At the feature learning layer, the model constructs three complementary base learners, a RoBERTa-FC for extracting global semantic features, a BERT-BiGRU for capturing temporal dependencies, and a TextCNN-Attention for focusing on local semantic features, thereby achieving multi-level text representation. At the decision layer, a meta-learner is used to fuse multi-view features, and dynamic uncertainty weighting and attention weighting strategies are employed to adaptively adjust outputs from different base learners. Experimental results across multiple domains demonstrate that the proposed model consistently outperforms single learners and comparison methods in terms of Accuracy, Precision, Recall, F1 Score, and Macro-AUC. On average, the ensemble model achieves a Macro-AUC of 0.9582 ± 0.023 across five datasets, with an Accuracy of 0.9423, an F1 Score of 0.9590, and a Macro-AUC of 0.9797 on the AlY_ds dataset. Moreover, in cross-dataset ranking evaluation based on equally weighted metrics, the model consistently ranks within the top two, confirming its superior cross-domain adaptability and robustness. These findings highlight the effectiveness of the proposed framework in enhancing sentiment classification performance and provide valuable insights for future research on lightweight dynamic ensembles, multilingual, and multimodal applications. Full article

This study aims to reconstruct NDI45 missing values due to cloud cover while outlining the importance of vegetation health for the climate–carbon cycle and the benefits of the NDI45 index for high canopy area indices. The methods include a novel hybrid framework that combines a deterministic Kalman filter (KF) and a clustering-based LSTM network to generate gap-free NDI45 series with 20 m spatial and 5-day temporal resolution. The innovation of the applied method relies on achieving a single-sensor workflow, provides a pixel-level uncertainty map, and minimizes LSTM overfitting through clustering based on a correlation threshold. In the northern Pampas (South America), this hybrid approach reduces the MAE by 22–35% on average and narrows the 95% confidence interval by 25–40% compared to the Kalman filter or LSTM alone. The three-dimensional spatio-temporal analysis demonstrates that the KF–LSTM hybrid provides better spatial homogeneity and reliability across the entire study area. The proposed framework can generate gap-free, high-resolution NDI45 time series with quantified uncertainties, enabling more reliable detection of vegetation stress, yield fluctuations, and long-term resilience trends. These capabilities make the method directly applicable to operational drought monitoring, crop insurance modeling, and climate risk assessment in agricultural systems, particularly in regions prone to frequent cloud cover. The framework can be further extended by including radar backscatter and multi-model ensembles, thus providing a promising basis for the reconstruction of global, high-resolution vegetation time series. Full article

A polyelectrolyte (PE) chain in the vicinity of an oppositely charged surface can exhibit a discontinuous transition from the adsorbed to the desorbed state once the electrostatic attractive interactions are not strong enough to overcome the entropic losses caused by the PE-surface adsorption. In the context of PE–protein interactions, the heterogeneity of the charge distribution and the effects of a low dielectric permittivity underneath the surface are crucial. Studies of the combined effects of these two properties are very sparse, especially in the spherical geometry; we thus fill this gap here. We study the adsorption of PE chains onto spherical particles with heterogeneously charged surfaces, with the main focus on the critical-adsorption conditions and the effects of a low-dielectric core. Metropolis Monte Carlo simulations are employed, with the PE exploring the phase-space around the binding particle in the canonical ensemble. Two adsorption–desorption transitions are observed when the particle possesses a net charge of the same sign as that of the PE, resulting in nonmonotonic behavior of the critical charge density required for the PE–particle electrostatically driven adsorption. An increased affinity between the PEs and low-dielectric particles with variable heterogeneous charge distributions is observed, in contrast to the behavior detected for homogeneous low-dielectric particles. This higher affinity occurs when the Debye screening length in the solution becomes comparable to the dimensions of a patch of the opposite sign to the PE. A number of real-life applications of the considered PE–particle system is presented in the discussion, in particular regarding the properties of the complex formation between various PEs and globular proteins featuring a dipolar-type distribution of electric charges on their surfaces, such as insulin and bovine serum albumin. Full article

To address overfitting due to limited sample size, and the challenges posed by “Spectral Homogeneity with Material Heterogeneity (SHMH)” and “Material Consistency with Spectral Divergence (MCSD)”—which arise from subtle spectral differences and limited classification accuracy—this study proposes a deep integration model that combines the Adaptive Boosting (AdaBoost) algorithm with a convolutional recurrent neural network (CRNN). The model adopts a dual-branch architecture integrating a 2D-CNN and gated recurrent unit to effectively fuse spatial and spectral features of rock samples, while the integration of the AdaBoost algorithm optimizes performance by enhancing system stability and generalization capability. The experiment used a hyperspectral dataset containing 81 rock samples (46 igneous rocks and 35 metamorphic rocks) and evaluated model performance through five-fold cross-validation. The results showed that the proposed 2D-CRNN-AdaBoost model achieved 92.55% overall accuracy, which was significantly better than that of other comparative models, demonstrating the effectiveness of multimodal feature fusion and ensemble learning strategy. Full article

While personal thermal comfort is critical for well-being and productivity, it is often overlooked by traditional building management systems that rely on uniform settings. Modern data-driven approaches often fail to capture the complex interactions between various data streams. This pilot study introduces a high-accuracy, interpretable framework for thermal comfort classification, designed to identify the most significant predictors from a comprehensive suite of environmental, physiological, and anthropometric data in a controlled group of young adults. Initially, an XGBoost model using the full 24-feature dataset achieved the best performance at 91% accuracy. However, after using SHAP analysis to identify and select the most influential features, the performance of our ensemble models improved significantly; notably, a Random Forest model’s accuracy rose from 90% to 94%. Our analysis confirmed that for this homogeneous cohort, environmental parameters—specifically temperature, humidity, and CO₂—were the dominant predictors of thermal comfort. The primary strength of this methodology lies in its ability to create a transparent pipeline that objectively identifies the most critical comfort drivers for a given population, forming a crucial evidence base for model design. The analysis also revealed that the predictive value of heart rate variability (HRV) diminished when richer physiological data, such as diastolic blood pressure, were included. For final validation, the optimized Random Forest model, using only the top 10 features, was tested on a hold-out set of 100 samples, achieving a final accuracy of 95% and an F1-score of 0.939, with all misclassifications occurring only between adjacent comfort levels. These findings establish a validated methodology for creating effective, context-aware comfort models that can be embedded into intelligent building management systems. Such adaptive systems enable a shift from static climate control to dynamic, user-centric environments, laying the critical groundwork for future personalized systems while enhancing occupant well-being and offering significant energy savings. Full article

Major depressive disorder is a mental illness characterized by persistent sadness or loss of interest that affects a person’s daily life. Early detection of this disorder is crucial for providing timely and effective treatment. Neuroimaging modalities, namely, functional magnetic resonance imaging, can be used to identify changes in brain regions related to major depressive disorder. In this study, regional homogeneity images, one of the derivative of functional magnetic resonance imaging is employed to detect major depressive disorder using the proposed feature/sample evolving voting ensemble approach. A total of 2380 subjects consisting of 1104 healthy controls and 1276 patients with major depressive disorder from Rest-meta-MDD consortium are studied. Regional homogeneity features from 90 regions are extracted using automated anatomical labeling template. These regional homogeneity features are then fed as an input to the proposed feature/sample selective evolving voting ensemble for classification. The proposed approach achieves an accuracy of 91.93%, and discriminative features obtained from the classifier are used to identify brain regions which may be responsible for major depressive disorder. A total of nine brain regions, namely, left superior temporal gyrus, left postcentral gyrus, left anterior cingulate gyrus, right inferior parietal lobule, right superior medial frontal gyrus, left lingual gyrus, right putamen, left fusiform gyrus, and left middle temporal gyrus, are identified. This study clearly indicates that these brain regions play a critical role in detecting major depressive disorder. Full article

Rapid population growth and urbanization have greatly impacted the environment, causing a sharp rise in city temperatures—a phenomenon known as the Urban Heat Island (UHI) effect. While previous research has extensively examined the influence of land use characteristics on urban heat islands, their impact on community demographics and UHI severity remains unexplored. Moreover, most previous studies have focused on specific locations, resulting in relatively homogeneous environmental data and limiting understanding of variations across different areas. To address this gap, this paper develops ensemble learning models to predict UHI severity based on demographic, meteorological, and land use/land cover factors in Midwestern United States. Analyzing over 11,000 data points from urban census tracts across more than 12 states in the Midwestern United States, this study developed Random Forest and XGBoost classifiers achieving weighted F1-scores up to 0.76 and excellent discriminatory power (ROC-AUC > 0.90). Feature importance analysis, supported by a detailed SHAP (SHapley Additive exPlanations) interpretation, revealed that the difference in vegetation between urban and rural areas (DelNDVI_summer) and imperviousness were the most critical predictors of UHI severity. This work provides a robust, large-scale predictive tool that helps urban planners and policymakers identify key UHI drivers and develop targeted mitigation strategies. Full article

The massive integration of variable renewable energy sources (RESs) poses the gradual necessity for new power system architectures with wide implementation of distributed battery energy storage systems (BESSs), which support power system stability, energy management, and control. This research presents a methodology and realization of a set of 11 BESS models based on different machine learning methods. The performance of the proposed models is tested using real-life BESS data, after which a comparative evaluation is presented. Based on the results achieved, a valuable discussion and conclusions about the models’ performance are made. This study compares the results of feedforward neural networks (FNNs), a homogeneous ensemble of FNNs, multiple linear regression, multiple linear regression with polynomial features, decision-tree-based models like XGBoost, CatBoost, and LightGBM, and heterogeneous ensembles of decision tree modes in the day-ahead forecasting of an existing real-life BESS in a PV power plant. A Bayesian hyperparameter search is proposed and implemented for all of the included models. Among the main objectives of this study is to propose hyperparameter optimization for the included models, research the optimal training period for the available data, and find the best model from the ones included in the study. Additional objectives are to compare the test results of heterogeneous and homogeneous ensembles, and grid search vs. Bayesian hyperparameter optimizations. Also, as part of the deep learning FNN analysis study, a customized early stopping function is introduced. The results show that the heterogeneous ensemble model with three decision trees and linear regression as main model achieves the highest average R² of 0.792 and the second-best nRMSE of 0.669% using a 30-day training period. CatBoost provides the best results, with an nRMSE of 0.662% for a 30-day training period, and offers competitive results for R²—0.772. This study underscores the significance of model selection and training period optimization for improving battery performance forecasting in energy management systems. The trained models or pipelines in this study could potentially serve as a foundation for transfer learning in future studies. Full article

This study presents a robust methodology for the indirect estimation of groundwater abstraction for irrigation at the scale of individual wells, addressing a key gap in data-scarce agricultural settings. The approach combines NDVI time series, crop water requirement modelling, and spatial analysis of irrigation systems within a GIS environment. A soil water balance model was applied to Homogeneous Units of Analysis, and irrigation requirements were estimated using an ensemble approach accounting for key sources of uncertainty related to phenology detection, soil moisture at sowing (%SAW), and irrigation system efficiency. A spatial linkage algorithm was developed to associate individual wells with the irrigated areas they supply. Sensitivity analysis demonstrated that 10% increases in %SAW resulted in abstraction reductions of up to 1.98%, while 10% increases in irrigation efficiency reduced abstractions by an average of 6.48%. These findings support the inclusion of both parameters in the ensemble, generating eight abstraction estimates per well. Values ranged from 33,000 to 115,000 m³ for the 2023 season. Validation against flowmeter data confirmed the method’s reliability, with an R² of 0.918 and an RMSE equivalent to 9.3% of the mean observations. This approach offers an accurate, spatially explicit estimation of groundwater abstractions without requiring direct metering and offers a transferable, cost-effective tool to improve groundwater accounting and governance in regions with limited monitoring infrastructure. Full article

Modeling of impurity diffusion processes in a multiphase randomly inhomogeneous body is performed using the Feynman diagram technique. The impurity diffusion equations are formulated for each of the phases separately. Their random boundaries are subject to non-ideal contact conditions for concentration. The contact mass transfer problem is reduced to a partial differential equation describing diffusion in the body as a whole, which accounts for jump discontinuities in the searched function as well as in its derivative at the stochastic interfaces. The obtained problem is transformed into an integro-differential equation involving a random kernel, whose solution is constructed as a Neumann series. Averaging over the ensemble of phase configurations is performed. The Feynman diagram technique is developed to investigate the processes described by parabolic partial differential equations. The mass operator kernel is constructed as a sum of strongly connected diagrams. An integro-differential Dyson equation is obtained for the concentration field. In the Bourret approximation, the Dyson equation is specified for a multiphase randomly inhomogeneous medium with uniform phase distribution. The problem solution, obtained using Feynman diagrams, is compared with the solutions of diffusion problems for a homogeneous layer, one having the coefficients of the base phase and the other having the characteristics averaged over the body volume. Full article

Current research lacks an in-depth exploration of ensemble learning and factor engineering applications in regard to landslide susceptibility modeling. In the Three Gorges Reservoir area of China, a region prone to frequent landslides that endanger lives and infrastructure, this study advances landslide susceptibility prediction by integrating ensemble learning with systematic factor engineering. Four homogeneous ensemble models (random forest, XGBoost, LightGBM, and CatBoost) and two heterogeneous ensembles (bagging and stacking) were implemented to evaluate 14 influencing factors. The key results demonstrate the Digital Elevation Model (DEM) as the dominant factor, while the stacking ensemble achieved superior performance (AUC = 0.876), outperforming single models by 4.4%. Iterative factor elimination and hyperparameter tuning increased the high-susceptibility zones in the stacking predictions to 42.54% and enhanced XGBoost’s low-susceptibility classification accuracy from 12.96% to 13.57%. The optimized models were used to generate a high-resolution landslide susceptibility map, identifying 23.8% of the northern and central regions as high-susceptibility areas, compared to only 9.3% as eastern and southern low-susceptibility zones. This methodology improved the prediction accuracy by 12–18% in comparison to a single model, providing actionable insights for landslide risk mitigation. Full article

The aim to reduce health risks of workers related to inhalative exposure to potentially toxic dusts requires the selection of appropriate measures depending on the hazard classification of the dust-composing materials. Due to their biodurability, respirable carbon fibers and their fragments can impose such health risks but are currently lacking hazard classification. Here, a method is presented for fragmenting carbon fiber materials and enriching fibrous fragments to a level that is expected to allow differentiating between fiber and particle overload-related toxic effects. The method was applied to a commercial polyacrylonitrile-based carbon fiber. It was ground in an oscillating ball mill, homogenized in a liquid using ultrasonication and left undisturbed for gravitational settling. This way, a vertical gradient in particle size and shape formed, from which the supernatant was collected. Fragment morphologies were characterized with large ensemble statistics by semi-automated evaluation of scanning electron microscopy images employing an artificial neural network for binary semantic segmentation. The number of fibrous fragments of respirable and thus critical fiber morphology was increased from $0.36\times {10}^6$ to $6\times {10}^6$ WHO-analog fibers per mg. This corresponds to a factor of about 15 compared to the initial ground material. Since the mass percentage of non-fibrous objects was also significantly reduced, the requirements for a subsequently scheduled toxicological study with intraperitoneal application were fulfilled. Intraperitoneal testing is an accepted method for assessing the carcinogenic potential of biopersistent fibers. The developed method allows enriching fibrous fractions of concern at acceptable throughput and enables testing fiber toxicological effects of respirable fragments from disintegrated polyacrylonitrile-based carbon fibers. Full article

Vertebral fracture (VF) may induce spinal cord injury that can lead to serious consequences which eventually may paralyze the entire or some parts of the body depending on the location and severity of the injury. Diagnosis of VFs is crucial at the initial stage, which may be challenging because of the subtle features, noise, and homogeneity present in the computed tomography (CT) images. In this study, Wide ResNet-40, DenseNet-121, and EfficientNet-B7 are chosen, fine-tuned, and used as base models, and a Bayesian-based probabilistic ensemble learning method is proposed for fracture detection in cervical spine CT images. The proposed method considers the prediction’s uncertainty of the base models and combines the predictions obtained from them, to improve the overall performance significantly. This method assigns weights to the base learners, based on their performance and confidence about the prediction. To increase the robustness of the proposed model, custom data augmentation techniques are performed in the preprocessing step. This work utilizes 15,123 CT images from the RSNA-2022 C-spine fracture detection challenge and demonstrates superior performance compared to the individual base learners, and the other existing conventional ensemble methods. The proposed model also outperforms the best state-of-the-art (SOTA) model by 1.62%, 0.51%, and 1.29% in terms of accuracy, specificity, and sensitivity, respectively; furthermore, the AUC score of the best SOTA model is lagging by 5%. The overall accuracy, specificity, sensitivity, and F1-score of the proposed model are 94.62%, 93.51%, 95.29%, and 93.16%, respectively. Full article

The strategic significance of software testing in ensuring the success of software development projects is paramount. Comprehensive testing, conducted early and consistently across the development lifecycle, is vital for mitigating defects, especially given the constraints on time, budget, and other resources often faced by development teams. Software defect prediction (SDP) serves as a proactive approach to identifying software components that are most likely to be defective. By predicting these high-risk modules, teams can prioritize thorough testing and inspection, thereby preventing defects from escalating to later stages where resolution becomes more resource intensive. SDP models must be continuously refined to improve predictive accuracy and performance. This involves integrating clean and preprocessed datasets, leveraging advanced machine learning (ML) methods, and optimizing key metrics. Statistical-based and traditional ML approaches have been widely explored for SDP. However, statistical-based models often struggle with scalability and robustness, while conventional ML models face challenges with imbalanced datasets, limiting their prediction efficacy. In this study, innovative decision forest (DF) models were developed to address these limitations. Specifically, this study evaluates the cost-sensitive forest (CS-Forest), forest penalizing attributes (FPA), and functional trees (FT) as DF models. These models were further enhanced using homogeneous ensemble techniques, such as bagging and boosting techniques. The experimental analysis on benchmark SDP datasets demonstrates that the proposed DF models effectively handle class imbalance, accurately distinguishing between defective and non-defective modules. Compared to baseline and state-of-the-art ML and deep learning (DL) methods, the suggested DF models exhibit superior prediction performance and offer scalable solutions for SDP. Consequently, the application of DF-based models is recommended for advancing defect prediction in software engineering and similar ML domains. Full article