Search Results (1,444)

Background: Cancer is a leading cause of death worldwide, and its early detection is crucial for improving patient outcomes. This study aimed to develop and evaluate ensemble learning models, specifically stacking, for the accurate prediction of lung, breast, and cervical cancers using lifestyle and clinical data. Methods: 12 base learners were trained on datasets for lung, breast, and cervical cancer. Stacking ensemble models were then developed using these base learners. The models were evaluated for accuracy, precision, recall, F1-score, AUC-ROC, MCC, and kappa. An explainable AI technique, SHAP, was used to interpret model predictions. Results: The stacking ensemble model outperformed individual base learners across all three cancer types. On average, for three cancer datasets, it achieved 99.28% accuracy, 99.55% precision, 97.56% recall, and 98.49% F1-score. A similar high performance was observed in terms of AUC, Kappa, and MCC. The SHAP analysis revealed the most influential features for each cancer type, e.g., fatigue and alcohol consumption for lung cancer, worst concave points, mean concave points, and worst perimeter for breast cancer and Schiller test for cervical cancer. Conclusions: The stacking-based multi-cancer prediction model demonstrated superior accuracy and interpretability compared with traditional models. Combining diverse base learners with explainable AI offers predictive power and transparency in clinical applications. Key demographic and clinical features driving cancer risk were also identified. Further research should validate the model on more diverse populations and cancer types. Full article

Ship type (e.g., Cargo, Tanker and Fishing) classification is crucial for marine management, environmental protection, and maritime safety, as it enhances navigation safety and aids regulatory agencies in combating illegal activities. Traditional ship type classification methods with AIS data are often plagued by problems such as data imbalance, insufficient feature extraction, reliance on single-model approaches, or unscientific model combination methods, which reduce the accuracy of classification. In this paper, we propose an ensemble classification method based on a stacking strategy to overcome these challenges. We apply the SMOTE technique to balance the dataset by generating minority class samples. Then, a more comprehensive ship behavior model is developed by combining static and dynamic features. A stacking strategy is adopted for the classification, integrating multiple tree structure-based classifiers to improve classification performance. The experimental results show that the ensemble classification method based on the stacking strategy outperforms traditional classifiers such as CatBoost, Random Forest, Decision Tree, LightGBM, and the ensemble classification method, especially in terms of improving classification precision, recall, F1 score, ROC curve, and AUC. This method improves the accuracy of ship type recognition, and it is suitable to real-time online classification, which is helpful for applications in marine safety monitoring, law enforcement, and illegal fishing detection. Full article

(1) Background: The harvest index is important for measuring the correlation between grain yield and aboveground biomass. However, the harvest index can only be measured after a mature harvest. If it can be obtained in advance during the growth period, it will promote research on high harvest indices and variety breeding; (2) Methods: In this study, we proposed a method to predict the harvest index during the rice growth period based on uncrewed aerial vehicle (UAV) remote sensing technology. UAV obtained visible light and multispectral images of different varieties, and the data such as digital surface elevation, visible light reflectance, and multispectral reflectance were extracted after processing for correlation analysis. Additionally, characteristic variables significantly correlated with the harvest index were screened out; (3) Results: The results showed that TCARI (correlation coefficient −0.82), GRVI (correlation coefficient −0.74), MTCI (correlation coefficient 0.83), and TO (correlation coefficient −0.72) had a strong correlation with the harvest index. Based on the above characteristics, this study used a variety of machine learning algorithms to construct a harvest index prediction model. The results showed that the Stacking model performed best in predicting the harvest index (R² reached 0.88) and had a high prediction accuracy. (4) Conclusions: Therefore, the harvest index can be accurately predicted during rice growth through UAV remote sensing images and machine learning technology. This study provides a new technical means for screening high harvest index in rice breeding, provides an important reference for crop management and variety improvement in precision agriculture, and has high application potential. Full article

Non-visual techniques for identifying sleep postures have become essential for enhancing sleep health. Conventional methods depend on a costly professional medical apparatus that is challenging to adapt for domestic use. This study developed an economical airbag mattress and introduced a method for detecting sleeping positions via restricted body pressure data. The methodology relies on distributed body pressure data obtained from barometric pressure sensors positioned at various locations on the mattress. Two combinations of base learners were chosen based on the complementary attributes of the model, each of which can be amalgamated through a soft-voting strategy. Additionally, the architectures of Autoencoder and convolutional neural networks were integrated, collectively constituting the base learning layer of the model. Gradient enhancement was utilized in the meta-learner layer to amalgamate the output of the basic learning layer. The experimental findings indicate that the suggested holistic learning model has high classification accuracy of up to 95.95%, precision of up to 96.13%, and F1 index of up to 95.01% in sleep posture recognition assessments and possesses considerable merit. In the subsequent application, the sleep monitoring device identified the sleep posture and employed an air conditioner and an air purifier to create a more comfortable sleep environment. The user can utilize the sleep posture data to improve the quality of sleep and prevent related diseases. Full article

The distribution of moisture content in stacked tea leaves during processing significantly influences tea quality. Visualizing this moisture distribution is crucial for optimizing processing parameters. In this study, we utilized hyperspectral imaging (HSI) technology combined with machine learning algorithms to evaluate the moisture content and its distribution in the stacked tea leaves in West Lake Longjing and Tencha green tea products during the processing flow line. A spectral quantitative determination model was developed, achieving high accuracy (${\mathrm{R}}_{\mathrm{p}}^2$ > 0.940) The model demonstrated strong generalization ability, allowing it to predict moisture content in both types of tea. Through hyperspectral imaging, we visualized moisture distribution in seven key processing steps and observed that moisture content was non-uniform, with the leaf tips and petioles having higher moisture levels than the leaf surface. This study offers a novel solution for real-time moisture monitoring of stacked tea leaves in tea production, ensuring consistent product quality. Future research could focus on refining model transfer techniques and exploring additional tea varieties to further enhance the generalization of the approach. Full article

Soil moisture (SM) profoundly influences crop growth, yield, soil temperature regulation, and ecological balance maintenance and plays a pivotal role in water resources management and regulation. The focal objective of this investigation is to identify feature parameters closely associated with soil moisture through the implementation of feature selection methods on multi-source remote sensing data. Specifically, three feature selection methods, namely SHApley Additive exPlanations (SHAP), information gain (Info-gain), and Info_gain ∩ SHAP were validated in this study. The multi-source remote sensing data collected from Sentinel-1, Landsat-8, and Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Digital Elevation Model (ASTGTM DEM) enabled the derivation of 25 characteristic parameters through sound computational approaches. Subsequently, a stacking algorithm integrating multiple machine-learning (ML) algorithms based on adaptive learning was engineered to accomplish soil moisture prediction. The attained prediction outcomes were then juxtaposed against those of single models, including Random Forest (RF), Adaptive Boosting (AdaBoost), Gradient Boosting Decision Tree (GBDT), Light Gradient Boosting Machine (LightGBM), Extreme Gradient Boosting (XGBoost), and Categorical Boosting (CatBoost). Notably, the adoption of feature factors selected by the Info_gain algorithm in combination with the adaptive stacking (Ada-Stacking) algorithm yielded the most optimal soil moisture prediction results. Specifically, the Mean Absolute Error (MAE) was determined to be 1.86 Vol. %, the Root Mean Square Error (RMSE) amounted to 2.68 Vol. %, and the R-squared (R2) reached 0.95. The multifactor integrated model that harnessed optical remote sensing data, radar backscatter coefficients, and topographic data exhibited remarkable accuracy in soil surface moisture retrieval, thus providing valuable insights for soil moisture inversion studies in the designated study area. Furthermore, the Ada-Stacking algorithm demonstrated its potency in integrating multiple models, thereby elevating retrieval accuracy and overcoming the limitations inherent in a single ML model. Full article

Structural health monitoring (SHM) plays a critical role in ensuring the safety and longevity of civil infrastructure by enabling the early detection of structural changes and supporting preventive maintenance strategies. In recent years, deep learning techniques have emerged as powerful tools for analyzing the complex data generated by SHM systems. This study investigates the use of deep belief networks (DBNs) for classifying structural conditions before and after retrofitting, using both ambient and train-induced acceleration data. Dimensionality reduction techniques such as principal component analysis (PCA) and t-distributed stochastic neighbor embedding (t-SNE) enabled a clear separation between structural states, emphasizing the DBN’s ability to capture relevant classification features. The DBN architecture, based on stacked restricted Boltzmann machines (RBMs) and supervised fine-tuning, was optimized via grid search and cross-validation. Compared to traditional unsupervised methods like K-means and PCA, DBNs demonstrated a superior performance in feature representation and classification accuracy. Experimental results showed median cross-validation accuracies of $98.04\%$ for ambient data and $96.96\%$ for train-induced data, with low variability. Although random forests slightly outperformed DBNs in classifying ambient data ($99.19\%$), DBNs achieved better results with more complex train-induced signals ($95.91\%$). Robustness analysis under Gaussian noise further demonstrated the DBN’s resilience, maintaining over $90\%$ accuracy for ambient data at noise levels up to ${\sigma }_{\mathrm{noise}}=0.5$. These findings confirm that DBNs are a reliable and effective approach for data-driven structural condition assessment in SHM systems. Full article

This study explores advanced process monitoring for one-shot drilling of aeronautical stacks made of aluminium 2024 and carbon fibre-reinforced polymer (CFRP) laminates using a 4.8 mm diameter drilling tool and unsupervised machine learning techniques. An experimental campaign is conducted to collect thrust force and torque signals at a 10 kHz sampling rate during the drilling process. These signals are employed for real-time process monitoring, focusing on material change detection and anomaly identification, where anomalies are defined as holes that fail to meet predefined quality criteria. An innovative approach based on unsupervised learning is proposed to enable automatic material change identification, signal segmentation, feature extraction, and hole quality assessment. Specifically, a semi-supervised approach based on a Gaussian Mixture Model (GMM) and 1D Convolutional AutoEncoder (1D-CAE) is employed to detect deviations from normal drilling conditions. The proposed method is benchmarked against state-of-the-art supervised techniques, including logistic regression (LR) and Support Vector Machines (SVMs). Results show that these traditional models struggle with class imbalance, leading to overfitting and limited generalisation, as reflected by the F1 scores of 0.78 and 0.75 for LR and SVM, respectively. In contrast, the proposed semi-supervised approach improves anomaly detection, achieving an F1 score of 0.87 by more effectively identifying poor-quality holes. This study demonstrates the potential of deep learning-based semi-supervised methods for intelligent process monitoring, enabling adaptive control in the drilling process of hybrid stacks and detecting anomalous holes. While the proposed approach effectively handles small and imbalanced datasets, further research into the application of generative AI could enhance performance, aiming for F1 scores above 0.90, thereby supporting adaptation in real industrial environments with high performance. Full article

The escalating global incidence of malware presents critical cybersecurity threats to manufacturing, automation, and industrial process control systems. Given the fast-developing web applications and IoT devices in use by industry operations, securing a transparent and effective malware detection mechanism has become imperative to operational resilience and data integrity. Classical methods of malware detection are conventionally opaque “black boxes” with limited transparency, thus eroding trust and hindering deployment in security-sensitive contexts. In this respect, this research proposes MAL-XSEL—a malware detection framework using an explainable stacking ensemble learning approach for performing high-accuracy classification and interpretable decision-making. MAL-XSEL explicates the model predictions through Shapley additive explanations (SHAP) and local interpretable model-agnostic explanations (LIME), which enable security analysts to validate how the detection logic works and prioritize the features contributing to the most critical threats. Evaluated on two benchmark datasets, MAL-XSEL outperformed conventional machine learning models, achieving top accuracies of 99.62% (ClaMP dataset) and 99.16% (MalwareDataSet). Notably, it surpassed state-of-the-art algorithms such as LightGBM (99.52%), random forest (99.33%), and decision trees (98.89%) across both datasets while maintaining computational efficiency. A unique interaction of ensemble learning and XAI is employed for detection, not only with improved accuracy but also with interpretable insight into the behavior of malware, thereby allowing trust to be substantiated in an automated system. By closing the divide between performance and interpretability, MAL-XSEL enables cybersecurity practitioners to deploy transparent and auditable defenses against an ever-growing resource of threats. This work demonstrates how there can be no compromise on explainability in security-critical applications and, as such, establishes a roadmap for future research on industrial malware analysis tools. Full article

To enhance the management of NOx emissions from hybrid electric buses, this paper develops an instantaneous NOx emission prediction model for hybrid electric buses based on a two-layer stacking ensemble learning method. Seventeen parameters, including operational characteristic parameters of hybrid electric buses, engine operating parameters, and emission after-treatment device operating parameters are selected as input features for the model. The correlation analysis results indicate that the Pearson correlation coefficients of engine coolant temperature and selective catalytic reduction (SCR) after-treatment device temperature show a significant linear negative correlation with instantaneous NOx emission mass. The Mutual Information (MI) analysis reveals that engine intake air volume, SCR after-treatment device temperature and engine fuel consumption have strong nonlinear relationships with instantaneous NOx emission mass. The two-layer stacking ensemble learning model selects eXtreme Gradient Boosting (XGBoost), Random Forest (RF), and an optimized BP neural network as base learners, with a linear regression model as the meta-learner, effectively predicting the instantaneous NOx emission mass of hybrid electric buses. The evaluation metrics of the proposed model—mean absolute error, root mean square error, and coefficient of determination—are 0.0068, 0.0283, and 0.9559, respectively, demonstrating a significant advantage compared to other benchmark models. Full article

Recently, the vehicle routing problem with pickup and delivery (VRP-PD) has attracted increasing interest due to its widespread applications in real-life logistics and transportation. However, existing learning-based methods often fail to fully exploit hierarchical graph structures, leading to suboptimal performance. In this study, we propose a graph-driven deep reinforcement learning (GDRL) approach that employs an encoder–decoder framework to address this shortcoming. The encoder incorporates stacked graph convolution modules (GCMs) to aggregate neighborhood information via updated edge features, producing enriched node representations for subsequent decision-making. The single-head attention decoder then applies a computationally efficient compatibility layer to sequentially determine the next node to visit. Extensive experiments demonstrate that the proposed GDRL achieves superior performance over both heuristic and learning-based baselines, reducing route length by up to 5.81% across synthetic and real-world datasets. Furthermore, GDRL also exhibits strong generalization capability across diverse problem scales and node distributions, highlighting its potential for real-world deployment. Full article

Depression has become one of the most common mental illnesses, causing severe physical and mental harm. To clarify the impact of brain region segmentation on the detection accuracy of moderate-to-severe major depressive disorder (MDD) and identify the optimal brain region for detecting MDD using electroencephalography (EEG), this study compared eight traditional single-machine learning algorithms with a hybrid machine learning model based on a stacking ensemble technique. The hybrid model employed K-nearest neighbors (KNN), decision tree (DT), and Extreme Gradient Boosting (XGBoost) as base learners and used a DT as the meta-learner. Compared with traditional single methods, the hybrid approach significantly improved detection accuracy by leveraging the strengths of different algorithms. In addition, this study divided the brain regions into the left and right temporal lobes and extracted both linear and nonlinear features to comprehensively capture the complexity and dynamic behavior of EEG signals, enhancing the model’s ability to distinguish features across different brain regions. The experimental results showed that among the eight traditional machine learning methods, the KNN classifier achieved the highest detection accuracy of 96.97% in the left temporal lobe region. In contrast, the stacking hybrid learning model further increased the detection accuracy to 98.07%, significantly outperforming the single models. Moreover, the analysis of the brain region segmentation revealed that the left temporal lobe exhibited higher discriminative power in detecting MDD, highlighting its important role in the neurobiology of depression. This study provides a solid foundation for developing more efficient and portable methods for detecting depression, offering new perspectives and approaches for EEG-based MDD detection, and contributing to the improvement in objectivity and precision in depression diagnosis. Full article

Stacking fault energy (SFE) is a critical property governing deformation mechanisms and influencing the mechanical behavior of materials. This work presents a unified framework for understanding and predicting SFE based solely on an electronic structure representation. By integrating density of states (DOS) spectral data, dimensionality reduction techniques, and machine learning models, it was found that the SFE behavior is indeed represented within the electronic structure and that this information can be used to accelerate the prediction of SFE. In the first part of this study, we established quantitative relationships between electronic structure and microstructural features, linking chemistry to mechanical properties. Using principal component analysis (PCA) and uniform manifold approximation and projection (UMAP), we identified key features from high-resolution vector representation of DOS data and explored their correlation with SFE. The second part of this work focuses on the predictive modeling of SFE, where a machine learning model trained on UMAP-reduced features achieved high accuracy (R² = 0.86, MAE = 15.46 mJ/m²). To bridge length scales, we extended this methodology to predict SFE in alloy systems, leveraging single-element data to inform multi-element alloy design. We illustrate this approach with Cu-Zn alloys, where the framework enabled rapid screening of compositional space while capturing complex electronic structure interactions. The proposed framework accelerates alloy design by reducing reliance on costly experiments and ab initio calculations. Full article

To address the performance degradation in existing PM_2.5 prediction models caused by excessive complexity, poor spatiotemporal efficiency, and suboptimal parameter optimization, we employ stacking ensemble learning for feature weighting analysis and integrate the ant colony optimization (ACO) algorithm for model parameter optimization. Combining meteorological and collaborative pollutant data, a model (namely the stacking-ACO-LSTM model) with a much shorter consuming time than that of only long short-term memory (LSTM) networks suitable for PM_2.5 concentration prediction is established. It can effectively filter out feature variables with higher weights, thereby reducing the predictive power of the model. The prediction of hourly PM_2.5 concentration of the model is trained and tested using real-time monitoring data in Nanchang City from 2017 to 2019. The results show that the established stacking-ACO-LSTM model has high accuracy in predicting PM_2.5 concentration, and compared to the same model without considering time and space efficiency and defective parameter optimization, the mean square error (MSE) decreases by about 99.88%, and the coefficient of determination (R²) increases by about 2.39%. This study provides a new idea for predicting PM_2.5 concentration in cities. Full article

This study advances crime analysis methodologies in Maryland by leveraging sophisticated machine learning (ML) techniques designed to cater to the state’s varied urban, suburban, and rural contexts. Our research utilized an enhanced combination of machine learning models, including random forest, gradient boosting, XGBoost, extra trees, and advanced ensemble methods like stacking regressors. These models have been meticulously optimized to address the unique dynamics and demographic variations across Maryland, enhancing our capability to capture localized crime trends with high precision. Through the integration of a comprehensive dataset comprising five years of detailed police reports and multiple crime databases, we executed a rigorous spatial and temporal analysis to identify crime hotspots. The novelty of our methodology lies in its technical sophistication and contextual sensitivity, ensuring that the models are not only accurate but also highly adaptable to local variations. Our models’ performance was extensively validated across various train–test split ratios, utilizing R-squared and RMSE metrics to confirm their efficacy and reliability for practical applications. The findings from this study contribute significantly to the field by offering new insights into localized crime patterns and demonstrating how tailored, data-driven strategies can effectively enhance public safety. This research importantly bridges the gap between general analytical techniques and the bespoke solutions required for detailed crime pattern analysis, providing a crucial resource for policymakers and law enforcement agencies dedicated to developing precise, adaptive public safety strategies. Full article