Search Results (287)

Deep learning models for power quality (PQ) disturbance classification remain critically vulnerable to adversarial perturbations, with classification performance degrading severely under white-box attacks. Existing defenses address individual models in isolation and provide no mechanism for operators to assess whether the system is under attack or which classifier remains trustworthy. This paper proposes a two-stage framework that combines adversarial training with large language model (LLM) reasoning to improve both robustness and interpretability. In the first stage, four architecturally diverse classifiers, including a proposed Multi-Scale Temporal Attention Network (MSTAN), are evaluated under four adversarial attacks (FGSM, PGD, C&W, and UAP), and their failure patterns are recorded as structured vulnerability fingerprints. The ensemble is then retrained via adversarial training on mixed clean and perturbed signals. In the second stage, an LLM analyzes the ensemble predictions alongside the fingerprint knowledge base to perform attack detection, fingerprint-guided meta-classification, and operator-facing threat report generation. On a 17-class, 255,000-signal synthetic benchmark, adversarial training recovers FGSM and PGD accuracy from below 25% to the 53–78% range, with MSTAN achieving the highest post-training robustness (78.26% under FGSM, 65.41% under PGD). The LLM reasoning layer provides an additional 3.5–6.2 percentage point improvement over majority voting by selecting the most reliable ensemble member based on the inferred attack condition, and detects adversarial attacks with 87.6% overall accuracy. To our knowledge, this is the first integration of LLM-based ensemble reasoning into the PQ adversarial robustness pipeline and the first application of the C&W optimization attack to power quality signals. Full article

Ensemble learning has been widely adopted in fraud detection; however, conventional ensemble strategies rely on uniform or performance-based weighting schemes that treat explainability as a post hoc annotation rather than an architectural component. This study addresses the research goal of whether SHAP attribution values can serve as a principled, instance-specific weighting mechanism within an ensemble, thereby embedding interpretability directly into the aggregation process. A SHAP-Weighted Ensemble (SWE) framework is proposed in which the L2 norm of each base model’s SHAP attribution vector, computed at prediction time, is used to derive instance-specific voting weights via Softmax normalization. Three linear base learners (logistic regression, robust LR, calibrated linear SVM) are combined, with LinearSHAP providing exact attribution values. A comprehensive evaluation protocol was applied on a real-world vehicle insurance claims dataset, including bootstrap 95% confidence intervals, McNemar’s test, a three-way ablation study comparing equal weighting, SWE, and validation-AUC weighting, F1-optimal threshold selection, expected calibration error, and cost-sensitive evaluation under asymmetric misclassification costs. The central finding is that SWE achieves performance statistically comparable to both simpler baselines across all evaluated metrics (ROC-AUC = 0.774, 95% CI [0.681, 0.862]; F1 = 0.679, 95% CI [0.569, 0.774]; McNemar p = 1.000), while producing a transparent, per-claim weighting trace that equal-weight voting cannot provide. A KernelSHAP influence analysis conducted directly on the SWE confirms that SHAP-derived weights are substantially aligned with actual model influence ratios (LR: 1.05×, LR_R: 1.05×, SVM: 0.81×), validating the weighting mechanism empirically. An exploratory analysis of a seven-model equal-weight diagnostic ensemble reveals a negative correlation (r = −0.721, p = 0.067) between individual model performance and ensemble influence; a theoretically coherent finding that does not reach statistical significance at conventional thresholds. The primary contribution of SWE is architectural and interpretability-driven: it produces an auditable, instance-level model-weighting mechanism grounded in SHAP attribution theory, supporting regulatory accountability under GDPR Article 22 and the EU AI Act. Full article

Although Random Forest (RF) is a widely adopted ensemble method for classification, preserving diversity among decision trees and reducing the negative impact of underperforming trees remain major challenges. This paper proposes DWRF-MVC, a novel RF optimization framework that integrates DBSCAN clustering with a multi-metric performance-weighted voting mechanism. Using a dissimilarity metric derived from classic RF outcomes, the framework applies DBSCAN to cluster decision trees, selects top-performing representatives from each cluster, and retains noise points to maintain diversity. A weighted voting mechanism is then introduced to further improve model performance. Experimental results on 14 benchmark datasets show that DWRF-MVC achieves an average accuracy improvement of 7.45% over classical RF with a 91.79% reduction in the number of trees. Moreover, DWRF-MVC surpasses two cutting-edge methods by 1.82% and 3.83% in average accuracy, respectively. Full article

Decentralized Autonomous Organizations (DAOs) enable blockchain-based collective governance, yet existing studies often evaluate DAO governance through isolated mechanisms, particularly voting systems. This narrow view does not sufficiently explain recurring problems such as governance capture, weak accountability, inadequate safeguards, and inefficient resource allocation. This paper proposes a Layered Governance Coverage Model that conceptualizes DAO governance as a system of seven interdependent institutional functions spanning participation, agenda formation, collective choice, safeguards, execution, incentives, and meta-governance. The model uses a four-level strength scale to assess not only whether governance functions are present, but also how strongly they are institutionalized. It is empirically applied to thirty-seven active DAOs through evidence-based coding of publicly available governance artifacts. The results show that governance breadth does not necessarily imply governance maturity: collective choice and execution mechanisms are more developed than accountability, safeguards, and meta-governance. Beyond DAO-native settings, the paper positions governance maturity as a trust and resilience regime for blockchain-based IoT and AI infrastructures, where governance affects security, reliability, data integrity, and risk oversight. The paper discusses AI-enabled governance analytics as a support mechanism for monitoring governance activity, detecting anomalies, and improving governance observability. The proposed framework contributes a structured approach for evaluating and designing resilient governance architectures in DAOs and blockchain-based IoT/AI systems. Full article

The South-to-North Water Diversion Project (SNWDP) in China involves a vast number of resettlers with far-reaching impacts. As a crucial carrier of resettlers’ daily lives, the indoor thermal comfort of resettlement housing directly affects their physical and mental health. However, existing empirical and field studies have paid limited attention to the thermal comfort and thermal adaptation of the resettlers. This study focuses on resettlers of the SNWDP, employing a combination of questionnaires and on-site measurements to analyze thermal benchmarks and thermal adaptation behavior data. The study introduces the concept of relative deprivation theory from social psychology, compares the correlations between vertical and horizontal deprivation and thermal perception across different relocation modes, and validates the predictive performance of commonly used thermal comfort models. The results show that as the relocation distance increases, the summer indoor thermal neutral temperature rises sequentially, and both the sensitivity to temperature changes and the width of the comfort zone also increase. Regarding thermal adaptation behaviors, the short-distance group primarily relies on passive adjustments such as using electric fans and reducing clothing, while the long-distance group significantly shifts toward active mechanical cooling like air conditioning. The sense of relative deprivation has a significant impact on the thermal comfort of medium- and long-distance resettlers, and its correlation even exceeds that of physical factors such as air temperature and black globe temperature. Among all groups, the ePMV and ePTS models modified by the expectancy factor exhibit the best predictive performance, with the smallest average deviation from the actual Thermal Sensation Vote (TSV), making them the optimal evaluation models for indoor thermal comfort of resettlers in the SNWDP. The findings provide theoretical guidance for creating healthy and comfortable indoor thermal environments in resettlement areas and for the sustainable development of subsequent phases of the SNWDP. Full article

Depression screening from social media has increasingly benefited from transformer-based architectures; however, integrating symptom-level analysis with clinically grounded diagnostic screening remains challenging. This study proposes a unified two-phase framework for social media-based depression detection aligned with DSM-5 criteria. In Phase 1, transformer-based learning strategies—Single, Voting, Stacking, Bagging, and Boosting—are employed to perform symptom-level multi-class classification of depressive symptoms. In Phase 2, the predicted symptoms are aggregated over a 14-day observation window to enable DSM-5-oriented binary depression screening. To ensure a robust and consistent evaluation, eight preprocessing configurations (D1–D8) are incorporated into the framework. Experimental results demonstrate that Bagging achieves the highest performance in symptom-level classification (F1 = 0.9394), while Voting and Boosting yield superior performance in DSM-5-oriented screening (F1-Yes = 0.7273). The findings reveal that different learning mechanisms play distinct roles across diagnostic levels, with variance-reduction strategies enhancing symptom differentiation and consensus-based approaches improving recall in clinical screening. This study provides a structured and clinically aligned framework for social media-based depression detection, offering practical insights for developing robust and scalable mental health screening systems. Full article

Microseismic early warning for roof disaster in excavated coal roadways often suffers from low pertinence and a high false positive rate. This study establishes an intelligent early warning process based on unsupervised learning and a voting mechanism. True triaxial compression and drilling tests were conducted to characterize the acoustic emission responses of coal and rock during fracture. Using 720 h of field microseismic data from a high-gas mine in Shanxi, high-weight precursor features were extracted from time–frequency indicators. Kernel principal component analysis (KPCA) was used to optimize the indicator system, and 49 indicators with weights above 0.08 were selected as model inputs. Five unsupervised clustering algorithms were integrated to establish an ensemble decision-making early warning model. The results show that the model eliminates the drawbacks of single algorithms, achieves accurate roof disaster warning, and correctly distinguishes disaster events from non-disaster high-energy events. The false positive rate is zero on the 720 h field dataset, and the reliability of early warning is significantly improved. This study enhances the reliability of mine roof microseismic warning, enriches roof disaster prediction theories, provides a complete intelligent early warning process for mine roof disaster, and offers important references for deep mining dynamic disaster warning research. Full article

Speech-based non-invasive screening offers a cost-effective and scalable approach for the early detection of Alzheimer’s disease (AD). However, the clinical utility of deep learning models remains severely constrained by the scarcity of labeled speech data in low-resource languages, necessitating cross-lingual transfer learning. Conventional domain adaptation paradigms typically assume semantically consistent feature domains and focus heavily on aligning marginal distributions; however, they suffer catastrophic performance degradation when applied to cross-lingual pathologic speech. By analyzing disease-associated representation vectors within a self-supervised HuBERT space, we uncover a systematic mechanism driving this failure, a phenomenon we term cross-lingual polarity flip, where the direction of disease-relative-to-control feature offsets fundamentally reverses between languages. While prior multilingual studies have largely discarded such dimensional inconsistencies as ungeneralizable noise, a 500-round Monte Carlo stability analysis demonstrates that these flips occur in a highly stable, structural manner across 18.3% of top discriminative dimensions. Leveraging this insight, we introduce Monte Carlo Polarity Flip Calibration (MC-PFC), a few-shot framework designed to explicitly rectify flip orientations. Requiring only five labeled support samples per class from the target domain, MC-PFC robustly estimates direction flips via a separability-weighted ensemble voting mechanism. Evaluated on a strictly held-out Chinese blind test set, MC-PFC achieves an area under the receiver operating characteristic curve (AUC) of 0.871, recovering 99.5% of the performance achieved by a full in-domain trained upper bound (AUC = 0.875). Ablation experiments confirm that direction calibration yields a substantial +0.361 AUC gain, vastly outperforming standard distribution alignment (+0.081). This work establishes a data-efficient paradigm for cross-lingual medical analysis, shifting the clinical AI focus from discarding cross-lingual discrepancies to actively modeling and calibrating them. Full article

Electronic voting requires the simultaneous admission of only legitimate participants, ballot uniqueness, vote confidentiality, storage integrity, and result verifiability. Blockchain alone does not solve these problems, since ledger immutability does not guarantee anonymity, ballot correctness, or reduced trust concentration. The purpose of this work is to develop a parameterizable research framework for electronic voting scenarios with enhanced cryptographic protection, allowing the security level to be varied according to the requirements of a voting scenario. The main contribution of the work is a parameterizable research architecture for composing and experimentally comparing electronic voting configurations with different security and computational profiles. The cryptographic and audit mechanisms integrated into this architecture include blind-signature-based anonymous authorization, encrypted ballot submission, blockchain-style audit, receipt verification, homomorphic tally publication, and threshold-supported tally artifacts. These mechanisms are not proposed as new cryptographic primitives; rather, they are integrated into a reproducible prototype to study how their combination affects verifiability, privacy support, auditability, and computational cost. Compared with basic blockchain-based voting prototypes, this architecture explicitly separates security, privacy, and verifiability profiles and makes their computational cost observable. The implemented prototype is used as an experimental platform for analyzing supported security properties, threat modeling, and computational cost estimation. The results show that authentication, anonymous token issuance, and receipt verification maintain an almost constant cost at the studied scale, while the main cryptographic burden is associated with encrypted ballot submission and threshold-supported tally publication. The scientific novelty of the work lies in constructing a parameterizable architecture that integrates several cryptographic mechanisms and a blockchain audit layer into one reproducible research prototype. At the same time, the proposed approach retains prototype-level limitations associated with the absence of a full zero-knowledge proof stack, independently deployed threshold authorities, and coercion-resistance mechanisms. Full article

Precise forecasting of the initial productivity rates of infill wells is essential for the effective exploitation of offshore reservoirs characterized by high water-cut. However, conventional reservoir simulation and basic machine learning models often suffer from high computational complexity and low interpretability. This research introduces a hybrid data-driven framework that combines ensemble feature engineering with a random forest model optimized through the crayfish optimization algorithm. The primary controlling factors were identified through a majority voting mechanism involving five feature selection algorithms. Subsequently, the COA was utilized to optimize the parameters of the random forest algorithm to improve its predictive robustness. The proposed EFE-COA-RF model achieves a testing MAE of 6.831 and an R² of 0.954, outperforming standard machine learning models and other optimization-based variants. The complete training process requires approximately 10.8 min, whereas the prediction time for the testing set is approximately 0.03 s. These results demonstrate that the proposed framework provides an accurate, interpretable, and efficient tool for rapid productivity evaluation in mature offshore oilfields. Full article

Accurate identification of HMs contamination in mine tailings is essential for understanding pollution and supporting remediation. However, conventional laboratory monitoring is labor-intensive, time-consuming, and spatially discontinuous, while UAV hyperspectral inversion is limited by spectral inconsistency, unstable performance under small-sample conditions, and insufficient interpretability. Here, we developed an interpretable UAV–laboratory synergistic framework for Cd and Pb mapping in the Yitong open-pit mine. Forty site-level soil samples, composited from 200 subsamples, were linked with UAV hyperspectral observations. Direct Standardization was used to harmonize UAV and laboratory spectra. A weighted voting ensemble (RF, GBDT, and XGBoost) achieved the best performance (R² = 0.85), outperforming the individual models and showing slightly higher stability than CNN (R² = 0.84). Environmental covariates (pH, SOM, SMC) revealed distinct metal-specific prediction patterns: Cd was mainly associated with pH–SOM interactions, whereas Pb was more strongly associated with SOM–SMC coupling. SHAP and Grad-CAM identified sensitive spectral regions, with Cd linked to the 440–580 nm range and Pb to the 720–740 nm range. Overall, this study provides an integrated framework that combines spectral transfer correction, stable multi-model inversion, and mechanism-oriented interpretability for HMs monitoring in complex mining environments. Full article

In semi-supervised learning scenarios, the presence of limited labeled data and abundant unlabeled samples poses significant challenges to model robustness and generalization. Although the semi-supervised broad learning system (SSBLS) effectively exploits manifold structure through graph Laplacian regularization, its optimization is typically formulated under the mean square error (MSE) criterion, which is sensitive to noise and outliers. To address this limitation, this paper introduces the maximum mixture correntropy criterion (MMC) into the SSBLS framework and proposes a model termed M2C-SSBLS. By replacing the conventional MSE loss with a mixture correntropy-based objective, the proposed method enhances robustness against non-Gaussian noise and abnormal samples while preserving the computational efficiency and analytical solution property of the BLS. Furthermore, to improve representation diversity and reduce model variance, a multi-view ensemble extension, named EC-SSBLS, is proposed. This method constructs multiple feature views through a random feature subspace strategy, and independently trains an M2C-SSBLS base learner on each subspace. Finally, the predicted results of each view are fused through a voting mechanism. Experiments on benchmark UCI datasets under noise-free, 10% and 20% label noise settings demonstrate that the proposed M2C-SSBLS consistently outperforms conventional SSBLS and other advanced semi-supervised learning approaches. The ensemble extension EC-SSBLS further enhances performance, particularly in noisy environments, validating the effectiveness of combining MMC-based optimization with multi-view ensemble learning. Full article

Breast cancer screening is a fundamentally probabilistic diagnostic task that requires precise identification of complex imaging characteristics from diverse patient cohorts. Despite improvements in deep learning techniques, current automatic tools are typically trained on well-curated datasets and do not generalize to heterogeneous data, thereby limiting their application. This study aims to address these shortcomings by introducing a more effective and generalizable framework for breast cancer classification that focuses on the stability of features, the learning of complementary representations, and improved decision modeling. The proposed methodology incorporates stability-driven feature extraction (SDFE) with a multi-branch architecture that consists of EfficientNetV2 (Convolutional neural networks (CNNs)), EfficientFormer (Vision transformers (ViTs)), and multi-layer perceptron (MLP)-Mixer models to extract various feature representations. To improve non-linear decision boundaries, it uses a Kolmogorov–Arnold Network (KAN)-based classification head and selects the most credible prediction via an adaptive voting mechanism. This model is trained using patient-level splitting on the VinDr-Mammo dataset, evaluated using five-fold cross-validation, and subsequently externally validated on the CBIS-DDSM dataset. Experimental findings demonstrate the consistent performance of the proposed model, with accuracies of 94.5% in cross-validation, 93.3% on the VinDr-Mammo test set, and 94.6% on CBIS-DDSM, surpassing other recent state-of-the-art solutions. It demonstrates enhanced robustness and cross-dataset generalization, offering a scalable, consistent framework for breast cancer classification that supports the development of computer-aided diagnostic systems. Full article

This study provides a comprehensive evaluation of bagging ensemble models for financial time series (FTS) classification and addresses a gap in the literature regarding how bootstrap methods, ensemble sizes, voting mechanisms, and loss functions jointly influence model performance. The analysis evaluates decision tree (DT), logistic regression (LR), and multi-layer perceptron (MLP) ensemble models modified by six time series bootstrap methods, five ensemble sizes, and three voting mechanisms across six FTS data sets. The study also examines the influence of entropy- and profit-based loss functions within particle swarm (PSO) and quantum-inspired particle swarm (QPSO) optimization for weighted voting. The results show that LR-based ensembles provide the strongest overall performance and outperform ARIMA, DT, LR, MLP, and LSTM baseline models on both accuracy and profit metrics. Bootstrap effects are model specific. DT and MLP ensembles perform best under the Tukey bootstrap, while LR ensembles achieve strong results under the block bootstrap, the sub-sample bootstrap method, and the Tukey method, and remain the strongest performers across all bootstrap configurations. Optimized voting mechanisms yield clear improvements over equal-weight majority voting, with the profit loss function producing the most consistent gains. The findings also indicate that FTS classification problems exhibit an optimal range of ensemble sizes, as larger ensembles do not always improve performance. The study contributes a systematic assessment of ensemble design choices for FTS classification and highlights the importance of jointly considering bootstrap diversity, ensemble size, and voting strategy when developing ensemble models for financial applications. Full article

Teaching organic chemistry is also important for pharmacists to understand the synthesis and mechanism of action of organic drug molecules. Unfortunately, organic chemistry is considered one of the most difficult subjects. This impression affects students’ motivation. To provide students with a learning aid and hopefully boost their motivation, this pilot study tested the integration of 28 elevenies—a special form of short poem—during a semester in an organic chemistry lecture for pharmacists. An anonymous and voluntary questionnaire was conducted at the end of the lecture sessions to assess perceptions of the use of elevenies as a teaching tool. Overall, the student feedback on the implementation of elevenies was positive. In general, students felt (with nearly 94% agreement) that a wider variety of learning methods, such as elevenies, should be incorporated into university teaching. They found elevenies, a type of literature, suitable for summarizing content of organic chemistry, a natural science. The majority (about 65%) stated that they secretly looked forward to the presentation of the elevenies during the lecture, indicating an increase in motivation. In addition, 83% of the respondents wanted to adduce elevenies to repeat and learn the teaching material. However, only about 20% could imagine writing elevenies themselves as part of the learning process. With 94% approval, the respondents gave a clear vote to retain elevenies in future semesters. This suggests the students’ perception of elevenies as an educational tool. Their ease of use could certainly be extended to other subject areas, provided that the topics addressed are not too complex. Full article