Search Results (345)

This study combines three optimization algorithms, Tunicate Swarm Algorithm (TSA), Whale Optimization Algorithm (WOA), and Jellyfish Search Optimizer (JSO), with random forest (RF) to predict the shear bond strength of cable bolts under different types and grouting conditions. Based on the original dataset, a database of 860 samples was generated by introducing random noise around each data point. After establishing three hybrid models (RF-WOA, RF-JSO, RF-TSA) and training them, the obtained models were evaluated using six metrics: coefficient of determination (R²), root mean squared error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), variance account for (VAF), and A-20 index. The results indicate that the RF-JSO model exhibits superior performance compared to the other models. The RF-JSO model achieved an excellent performance on the testing set (R² = 0.981, RMSE = 11.063, MAE = 6.457, MAPE = 9, VAF = 98.168, A-20 = 0.891). In addition, Shapley Additive exPlanations (SHAP), Partial Dependence Plot (PDP), and Local Interpretable Model-agnostic Explanations (LIME) were used to analyze the interpretability of the model, and it was found that confining pressure (Stress), elastic modulus (E), and a standard cable type (cable type_standard) contributed the most to the prediction of shear bond strength. In summary, the hybrid model proposed in this study can effectively predict the shear bond strength of cable bolts. Full article

Land use efficiency (LUE) serves as a crucial nexus between economic development and sustainable resource management, directly influencing urban production–consumption systems. While economic development stages (EDSs) reflect a region’s environmental carrying capacity and profoundly affect LUE, the specific mechanisms governing this relationship remain unclear. In this study, we combined multi-source data to portray the spatiotemporal patterns of EDSs and LUE in 276 Chinese cities from 1995 to 2020, and we identified the nonlinear effects of EDSs on LUE. Based on the fine-scale LUE, it is confirmed that the older the age of urban land generation, the higher the LUE, laying a theoretical foundation for subsequent research. Simultaneously, the EDS continues to be upgraded, with approximately 70% of cities reaching the post-industrialization stage or higher by 2020. The results of partial dependency plots (PDPs) revealed that the EDS has a positive impact on LUE. From the perspective of different urban scales, the higher the EDSs of supercities, type I large cities, type II large cities, and type II small cities, the greater the positive impact on LUE, whereas the impact patterns at other urban scales follow an inverted U-shape. These findings carry important implications for sustainable spatial development, particularly in optimizing land resource allocation to assist the shift to more efficient production systems and responsible consumption patterns. Full article

The evolution toward sixth generation (6G) wireless networks promises higher performance, greater flexibility, and enhanced intelligence. However, it also introduces a substantially enlarged attack surface driven by open, disaggregated, and multi-vendor Open RAN (O-RAN) architectures that will be utilised in 6G networks. This paper addresses the urgent need for a practical Zero Trust (ZT) deployment model tailored to O-RAN specification. To do so, we introduce a novel hybrid ZT deployment model that establishes the trusted foundation for AI/ML-driven security in O-RAN, integrating macro-level enclave segmentation with micro-level application sandboxing for xApps/rApps. In our model, the Policy Decision Point (PDP) centrally manages dynamic policies, while distributed Policy Enforcement Points (PEPs) reside in logical enclaves, agents, and gateways to enable per-session, least-privilege access control across all O-RAN interfaces. We demonstrate feasibility via a Proof of Concept (PoC) implemented with Kubernetes and Istio and based on the NIST Policy Machine (PM). The PoC illustrates how pods can represent enclaves and sidecar proxies can embody combined agent/gateway functions. Performance discussion indicates that enclave-based deployment adds 1–10 ms of additional per-connection latency while CPU/memory overhead from running a sidecar proxy per enclave is approximately 5–10% extra utilisation, with each proxy consuming roughly 100–200 MB of RAM. Full article

Accurate maize yield forecasting under climate uncertainty remains a critical challenge for global food security, yet existing studies predominantly rely on single-model frameworks, limiting generalizability and actionable insights. This study selected three regions, specifically Dali, Lijiang, and Zhaotong, and collected data on 12 agronomic traits of 114 varieties, along with eight sets of meteorological data, covering the period from 2019 to 2023. We employed three machine learning models: Random Forest (RF), Support Vector Machine (SVM), and XGBoost. The results revealed a strong correlation between yield and multiple agronomic traits, particularly grain weight per spike (GWPS) and hundred-kernel weight (HKW). Notably, the XGBoost model emerged as the top performer across all three regions. The model achieved the lowest RMSE (0.22–191.13) and a good R² (0.98–0.99), demonstrating exceptional predictive accuracy for yield-related traits. The comparative analysis revealed that XGBoost exhibited superior accuracy and stability compared to RF and SVM. Through feature importance analysis, four critical determinants of yield were identified: GWPS, shelling percentage (SP), growth period (GP), and plant height (PH). Furthermore, partial dependence plots (PDPs) provided deeper insights into the nonlinear interactive effects between GWPS, SP, GP, PH, and yield, offering a more comprehensive understanding of their complex relationships. This study presents an innovative, data-driven methodology designed to accurately forecast corn yield across diverse locations. This approach offers valuable scientific insights that can significantly enhance precision agricultural practices by enabling the precise tailoring of fertilizer usage and irrigation strategies. The results highlight the importance of integrating agronomic and meteorological data in yield forecasting, paving the way for development of agricultural decision-support systems in the context of future climate change scenarios. This study presents an innovative, data-driven methodology designed to accurately forecast corn yield across diverse locations. This approach offers valuable scientific insights that can significantly enhance precision agricultural practices by enabling the precise tailoring of fertilizer usage and irrigation strategies. Full article

This paper presents a cross-layer intrusion detection framework leveraging explainable artificial intelligence (XAI) and interpretability methods to enhance transparency and robustness in attack detection within the Internet of Medical Things (IoMT) domain. By addressing the dual challenges of compromised data integrity, which span both biosensor and network-layer data, this study combines advanced techniques to enhance interpretability, accuracy, and trust. Unlike conventional flow-based intrusion detection systems that primarily rely on transport-layer statistics, the proposed framework operates directly on raw packet-level features and application-layer semantics, including MQTT message types, payload entropy, and topic structures. The key contributions of this research include the application of K-Means clustering combined with the principal component analysis (PCA) algorthim for initial categorization of attack types, the use of SHapley Additive exPlanations (SHAP) for feature prioritization to identify the most influential factors in model predictions, and the employment of Partial Dependence Plots (PDP) and Accumulated Local Effects (ALE) to elucidate feature interactions across layers. These methods enhance the system’s interpretability, making data-driven decisions more accessible to nontechnical stakeholders. Evaluation on a realistic healthcare IoMT testbed demonstrates significant improvements in detection accuracy and decision-making transparency. Furthermore, the proposed approach highlights the effectiveness of explainable and cross-layer intrusion detection for secure and trustworthy medical IoT environments that are tailored for cybersecurity analysts and healthcare stakeholders. Full article

The global acceleration of population urbanization has transformed cities into primary spatial hubs of human activity. As urban populations continue to expand, identifying the socioeconomic drivers of urbanization and elucidating their underlying mechanisms are essential for achieving Sustainable Development Goal 11, established by the United Nations. This study leverages machine learning and big data to investigate the determinants of population urbanization in China over the period 1991–2023. Utilizing the XGBoost algorithm combined with SHAP (Shapley Additive Explanations), the analysis reveals a tripartite structure of key drivers encompassing industrial support, employment orientation, and infrastructure accessibility. Regional assessments indicate distinct urbanization patterns: Eastern coastal areas are predominantly driven by finance and service industries; central inland regions follow an investment-led trajectory anchored in infrastructure development and real estate expansion, while the western interior relies mainly on employment-centered strategies. Partial Dependence Plots (PDPs) highlighted spatial variations in the effects of sensitive factors, with interaction analyses revealing synergistic effects between tertiary sector shares and the working-age share in eastern coastlands, structural amplification by real estate investment with appropriate working-age population shares in the central inlands, and balancing interactions between GDP growth rates and tertiary sector shares in the western interior. These findings contribute to a more nuanced understanding of the socioeconomic forces shaping urbanization and offer evidence-based recommendations for policymakers in other developing countries seeking to foster sustainable urban growth. Full article

This paper presents a comparative study of data-driven modeling approaches for vanadium redox flow batteries (VRFBs), utilizing Multiple Linear Regression (MLR) and Random Forest (RF) algorithms. Experimental voltage–capacity datasets from a 1 kW/1 kWh VRFB system were digitized, processed, and used for model training, validation, and testing. The MLR model, built using eight optimized features, achieved a mean error (ME) of 0.0204 V, a residual sum of squares (RSS) of 8.87, and a root mean squared error (RMSE) of 0.1796 V on the test data, demonstrating high predictive performance in stationary operating regions. However, it exhibited limited accuracy during dynamic transitions. Optimized through out-of-bag (OOB) error minimization, the Random Forest model achieved a training RMSE of 0.093 V and a test RMSE of 0.110 V, significantly outperforming MLR in capturing dynamic behavior while maintaining comparable performance in steady-state regions. The accuracy remained high even at lower current densities. Feature importance analysis and partial dependence plots (PDPs) confirmed the dominance of current-related features and SOC dynamics in influencing VRFB terminal voltage. Overall, the Random Forest model offers superior accuracy and robustness, making it highly suitable for real-time VRFB system monitoring, control, and digital twin integration. This study highlights the potential of combining machine learning algorithms with electrochemical domain knowledge to enhance battery system modeling for future energy storage applications. Full article

Understanding the carbon emissions of multimodal travel—comprising walking, metro, bus, cycling, and ride-hailing—is essential for promoting sustainable urban mobility. However, most existing studies focus on single-mode travel, while underlying spatiotemporal and behavioral determinants remain insufficiently explored due to the lack of fine-grained data and interpretable analytical frameworks. This study proposes a novel integration of high-frequency, real-world mobility trajectory data with interpretable machine learning to systematically identify the key drivers of carbon emissions at the individual trip level. Firstly, multimodal travel chains are reconstructed using continuous GPS trajectory data collected in Beijing. Secondly, a model based on Calculate Emissions from Road Transport (COPERT) is developed to quantify trip-level CO₂ emissions. Thirdly, four interpretable machine learning models based on gradient boosting—XGBoost, GBDT, LightGBM, and CatBoost—are trained using transportation and built environment features to model the relationship between CO₂ emissions and a set of explanatory variables; finally, Shapley Additive exPlanations (SHAP) and partial dependence plots (PDPs) are used to interpret the model outputs, revealing key determinants and their non-linear interaction effects. The results show that transportation-related features account for 75.1% of the explained variance in emissions, with bus usage being the most influential single factor (contributing 22.6%). Built environment features explain the remaining 24.9%. The PDP analysis reveals that substantial emission reductions occur only when the shares of bus, metro, and cycling surpass threshold levels of approximately 40%, 40%, and 30%, respectively. Additionally, travel carbon emissions are minimized when trip origins and destinations are located within a 10 to 11 km radius of the central business district (CBD). This study advances the field by establishing a scalable, interpretable, and behaviorally grounded framework to assess carbon emissions from multimodal travel, providing actionable insights for low-carbon transport planning and policy design. Full article

The corrosion behavior of an additively manufactured Mg-1Zn alloy was investigated in both the transverse and longitudinal directions relative to the build direction, in the as-built condition and after annealing at 350 °C for 24 h under high vacuum. Microstructural characterization using XRD and SEM revealed the presence of magnesium oxide (MgO) and the absence of intermetallic second-phase particles. Optical microscopy (OM) images and Electron Backscatter Diffraction (EBSD) maps showed a highly complex grain morphology with anomalous, anisotropic shapes and a heterogeneous grain size distribution. The microstructure includes grains with a pronounced columnar morphology aligned along the build direction and is therefore characterized by a strong crystallographic texture. Electrochemical techniques, including PDP and EIS, along with gravimetric H₂ collection, concluded that the transverse plane exhibited greater corrosion resistance compared to the longitudinal plane. Additionally, an increase in cathodic kinetics was observed when comparing as-built with heat-treated samples. Full article

Background: Overcrowded emergency departments (EDs) create significant challenges for patient management and hospital efficiency. In response, Amiens Picardy University Hospital (APUH) developed the “Prediction of the Patient Pathway in the Emergency Department” (3P-U) model to enhance patient flow management. Objectives: To develop and clinically validate an explainable artificial intelligence (XAI) model for hospital admission predictions, using structured triage data, and demonstrate its real-world applicability in the ED setting. Methods: Our retrospective, single-center study involved 351,019 patients consulting in APUH’s EDs between 2015 and 2018. Various models (including a cross-validation artificial neural network (ANN), a k-nearest neighbors (KNN) model, a logistic regression (LR) model, and a random forest (RF) model) were trained and assessed for performance with regard to the area under the receiver operating characteristic curve (AUROC). The best model was validated internally with a test set, and the F1 score was used to determine the best threshold for recall, precision, and accuracy. XAI techniques, such as Shapley additive explanations (SHAP) and partial dependence plots (PDP) were employed, and the clinical explanations were evaluated by emergency physicians. Results: The ANN gave the best performance during the training stage, with an AUROC of 83.1% (SD: 0.2%) for the test set; it surpassed the RF (AUROC: 71.6%, SD: 0.1%), KNN (AUROC: 67.2%, SD: 0.2%), and LR (AUROC: 71.5%, SD: 0.2%) models. In an internal validation, the ANN’s AUROC was 83.2%. The best F1 score (0.67) determined that 0.35 was the optimal threshold; the corresponding recall, precision, and accuracy were 75.7%, 59.7%, and 75.3%, respectively. The SHAP and PDP XAI techniques (as assessed by emergency physicians) highlighted patient age, heart rate, and presentation with multiple injuries as the features that most specifically influenced the admission from the ED to a hospital ward. These insights are being used in bed allocation and patient prioritization, directly improving ED operations. Conclusions: The 3P-U model demonstrates practical utility by reducing ED crowding and enhancing decision-making processes at APUH. Its transparency and physician validation foster trust, facilitating its adoption in clinical practice and offering a replicable framework for other hospitals to optimize patient flow. Full article

The increasing complexity of manufacturing processes demands accurate defect prediction and interpretable insights into the causes of quality issues. This study proposes a methodology integrating machine learning, clustering, and Explainable Artificial Intelligence (XAI) to support defect analysis and quality control in industrial environments. Using a dataset based on empirical industrial distributions, we train an XGBoost model to classify high- and low-defect scenarios from multidimensional production and quality metrics. The model demonstrates high predictive performance and is analyzed using five XAI techniques (SHAP, LIME, ELI5, PDP, and ICE) to identify the most influential variables linked to defective outcomes. In parallel, we apply Fuzzy C-Means and K-means to segment production data into latent operational profiles, which are also interpreted using XAI to uncover process-level patterns. This approach provides both global and local interpretability, revealing consistent variables across predictive and structural perspectives. After a thorough review, no prior studies have combined supervised learning, unsupervised clustering, and XAI within a unified framework for manufacturing defect analysis. The results demonstrate that this integration enables a transparent, data-driven understanding of production dynamics. The proposed hybrid approach supports the development of intelligent, explainable Industry 4.0 systems. Full article

Coatings that are tolerant of poor surface preparation are often used for rapid, real-time maintenance of aging steel surfaces. In this study, a modified epoxy (EP) anti-rust coating was proposed, utilizing methyl gallate (MG) as a rust conversion agent, graphene oxide (GO) as an active functional material, and epoxy resin as the film-forming material. The anti-rust mechanism was investigated using potentiodynamic polarization (PDP), electrochemical impedance spectroscopy (EIS), scanning electron microscopy (SEM), laser scanning confocal microscopy (LSCM), and the scanning vibration electrode technique (SVET). The results demonstrated that over a period of 21 days, the impedance of the coating increases while the corrosion current density decreases with prolonged soaking time. The coating exhibited a maximum impedance of 2259 kΩ, and a lower corrosion current density of 8.316 × 10⁻³ A/m², which demonstrated a three-order magnitude reduction compared to the corrosion current density observed in mild steel without coating. LSCM demonstrated that MG can not only penetrate the tiny gap between the rust particles, but also effectively convert harmful rust into a complex. SVET showed a much more uniform current density distribution in the micro-zones of mild steel with the anti-rust coating compared to uncoated mild steel, indicating that the presence of GO not only enhanced the electrical conductivity of the coating, but also improved the structure of the coating, which contributed to the high performance of the modified epoxy anti-rust coating. This work highlights the potential application of anti-rust coating in the protection of metal structures in coastal engineering. Full article

Geopolymers, as an eco-friendly alternative construction material to ordinary Portland cement (OPC), exhibit superior performance in soil stabilization. However, their inherent brittleness limits engineering applications. To address this, polyethylene terephthalate (PET) fibers can be incorporated into a one-part geopolymer (OPG) binder to enhance ductility while promoting plastic waste recycling. However, the evaluation of ductile behavior of OPG-stabilized soil with PET fiber normally demands extensive laboratory and field experiments. Leveraging artificial intelligence, a predictive model can be developed for this purpose. In this study, data were collected from compressive and tensile tests performed on the OPG-stabilized soil with PET fiber. Four deep learning neural network models, namely ANN, BPNN, CNN, and LSTM, were then used to construct prediction models. The input parameters in the model included the fly ash (FA) dosage, dosage and length of the PET fiber, and the Curing Time. Results revealed that the LSTM model had the best performance in predicting the three ductile properties (i.e., the compressive strength index [UCS], strain energy index [CSE], and tensile strength index [TES]). The SHAP and 2D-PDP methods were further used to verify the rationality of the LSTM model. It is found that the Curing Time was the most important factor for the strength and ductile behavior. The appropriate addition of PET fiber of a certain length had a positive impact on the ductility index. Thus, for the OPG-stabilized soil, the optimal dosage and length of PET fiber were found to be 1.5% and 9 mm, respectively. Additionally, there was a synergistic effect between FA and PET on the ductility metric. This research provides theoretical support for the application of geopolymer and PET fiber in enhancing the ductility of the stabilized soil. Full article

Liquid rocket engines occasionally experience abnormal phenomena with unclear mechanisms, causing difficulty in design improvements. To address the above issue, a data mining method that combines ante hoc explainability, post hoc explainability, and prediction accuracy is proposed. For ante hoc explainability, a feature selection method driven by data, models, and domain knowledge is established. Global sensitivity analysis of a physical model combined with expert knowledge and data correlation is utilized to establish the correlations between different types of parameters. Then a two-stage optimization approach is proposed to obtain the best feature subset and train the prediction model. For the post hoc explainability, the partial dependence plot (PDP) and SHapley Additive exPlanations (SHAP) analysis are used to discover complex patterns between input features and the dependent variable. The effectiveness of the hybrid feature selection method and its applicability under different noise combinations are validated using synthesized data from a high-fidelity simulation model of a pressurization system. Then the analysis of the causes of a large vibration phenomenon in an active engine shows that the prediction model has good accuracy, and the feature selection results have a clear mechanism and align with domain knowledge, providing both accuracy and interpretability. The proposed method shows significant potential for data mining in complex aerospace products. Full article

Provable data possession (PDP) is a technique that enables the verification of data integrity in cloud storage without the need to download the data. PDP schemes are generally categorized into public and private verification. Public verification allows third parties to assess the integrity of outsourced data, offering good openness and flexibility, but it may lead to privacy leakage and security risks. In contrast, private verification restricts the auditing capability to the data owner, providing better privacy protection but often resulting in higher verification costs and operational complexity due to limited local resources. Moreover, most existing PDP schemes are based on classical number-theoretic assumptions, making them vulnerable to quantum attacks. To address these challenges, this paper proposes an identity-based PDP with a designated verifier over lattices, utilizing a specially leveled identity-based fully homomorphic signature (IB-FHS) scheme. We provide a formal security proof of the proposed scheme under the small-integer solution (SIS) and learning with errors (LWE) within the random oracle model. Theoretical analysis confirms that the scheme achieves security guarantees while maintaining practical feasibility. Furthermore, simulation-based experiments show that for a 1 MB file and lattice dimension of n = 128, the computation times for core algorithms such as TagGen, GenProof, and CheckProof are approximately 20.76 s, 13.75 s, and 3.33 s, respectively. Compared to existing lattice-based PDP schemes, the proposed scheme introduces additional overhead due to the designated verifier mechanism; however, it achieves a well-balanced optimization among functionality, security, and efficiency. Full article