Epilepsy is a prevalent neurological disorder affecting approximately 50 million individuals worldwide. Electroencephalogram (EEG) signals play a vital role in the diagnosis and analysis of epileptic seizures. However, traditional machine learning techniques often rely on handcrafted features, limiting their robustness and generalizability across diverse EEG acquisition settings, seizure types, and patients. To address these limitations, we propose RDPNet, a multi-scale residual dilated pyramid network with entropy-guided feature fusion for automated epileptic EEG classification. RDPNet combines residual convolution modules to extract local features and a dilated convolutional pyramid to capture long-range temporal dependencies. A dual-pathway fusion strategy integrates pooled and entropy-based features from both shallow and deep branches, enabling robust representation of spatial saliency and statistical complexity. We evaluate RDPNet on two benchmark datasets: the University of Bonn and TUSZ. On the Bonn dataset, RDPNet achieves 99.56–100% accuracy in binary classification, 99.29–99.79% in ternary tasks, and 95.10% in five-class classification. On the clinically realistic TUSZ dataset, it reaches a weighted F1-score of 95.72% across seven seizure types. Compared with several baselines, RDPNet consistently outperforms existing approaches, demonstrating superior robustness, generalizability, and clinical potential for epileptic EEG analysis.
Accurate long-term residential load forecasting is critical for modern power grid scheduling and demand response. To address the challenges of non-stationary distribution shifts and the limitations of single-scale pattern capturing, this paper proposes RHFormer, an enhanced long-sequence load forecasting model built upon a deep decomposition framework. The model integrates Reversible Instance Normalization (RevIN) to mitigate the effects of non-stationarity and introduces a hybrid correlation module. By combining global auto-correlation with local multi-scale depthwise separable convolutions (MS-DSC), RHFormer enables collaborative modeling of complex load patterns. Experimental results on a public electricity dataset demonstrate that the proposed model consistently outperforms mainstream baselines across multiple forecasting horizons.
Disputes involving vulnerable groups in grassroots public governance exhibit diverse characteristics, with the concurrent trends of population mobility and aging further complicating dispute mediation. This study employs the four core elements of social vulnerability theory—exposure, sensitivity, adaptive capacity, and resilience—as an explanatory framework to systematically identify barriers in grassroots mediation processes and propose low-cost recommendations compatible with existing procedures. The research divides the mediation process into four stages: (1) intake and orientation, (2) dispute communication and focus, (3) mediation solution formulation and implementation, and (4) post-settlement follow-up and stabilization. It analyzes the specific manifestations of vulnerability elements across these stages. Results indicate that the four dimensions of vulnerability interact in complex ways throughout the mediation process, creating different types of participation barriers. To address issues related to exposure, sensitivity, adaptive capacity, and resilience, this study proposes targeted strategies, including optimizing information delivery methods, providing structured support tools, and establishing progressive implementation designs. These recommendations help enhance the inclusiveness and accessibility of mediation processes, ensuring that different groups can participate equally and make effective use of mediation mechanisms.
Background/Objectives: Motor imagery electroencephalography (MI-EEG) contains weak, non-stationary patterns that evolve across channels and time. Effective decoding requires local sensorimotor structure, trial-wide context, and ordered temporal evolution to be represented without discarding information from earlier processing stages. Methods: We propose STDR-Net, a Spatio-Temporal Dependency Refinement Network that recalibrates convolutional tokens, models their global interactions, restores the original local representation through a cross-stage residual bridge, and refines the fused sequence with a bidirectional long short-term memory network. STDR-Net was evaluated on BCI Competition IV-2a, BCI Competition IV-2b, and the PhysioNet EEG Motor Movement/Imagery dataset against nine reproduced baselines. The PhysioNet experiment used a pooled mixed-participant trial-level split rather than an unseen-participant evaluation. Results: STDR-Net obtained accuracies of 82.69%, 86.99%, and 59.76% and Cohen’s kappa values of 0.7692, 0.7398, and 0.46, respectively. It achieved the highest observed mean accuracy on IV-2a and competitive performance on IV-2b and PhysioNet. Bridge-specific, ordering, module, and augmentation ablations supported the proposed pathway on IV-2a, whereas the differences on IV-2b were small and not significant after multiplicity correction. Conclusions: The results support information-preserving dependency refinement as a competitive design for MI-EEG decoding, while also showing that its benefit depends on the dataset and evaluation setting.