Electricity demand forecasting plays a crucial role in energy planning and power system operation. However, it is affected by numerous factors and complex relationships, making accurate prediction challenging. Therefore, from the perspective of sample diversity in the base dataset, we propose an improved stacking-based ensemble regression algorithm to enhance the accuracy of electricity demand forecasting. Firstly, a continuous sampling strategy is constructed between the sample integration selection probability and the base dataset using D2-Sampling and KNN; secondly, multiple base regression models are integrated through stacking to improve the predictive performance. In the electricity demand forecasting experiments conducted on three different datasets and across multiple base models, the proposed improved stacking ensemble learning regression algorithm (DK-Stacking) achieved the best performance. This symmetric experimental evaluation ensured consistent and balanced assessment of the model performance across datasets and models, highlighting the robustness and generalization of the proposed algorithm. Compared to the ANN, SVR, and RF models, its prediction accuracy increased by more than 1 percentage point. Even when compared to the optimized XGBoost model, it showed an improvement of 0.44 percentage points. Overall, the proposed DK-Stacking demonstrates symmetry-inspired robustness in electricity demand forecasting through the balanced treatment of datasets and model integration.
Epilepsy EEG signal recognition is an important method for epilepsy detection. However, the traditional machine learning methods used for epilepsy EEG signal recognition face many challenges. For example, traditional intelligent recognition methods often ignore the imbalance of epilepsy EEG signals, which misclassifies positive samples and may cause serious consequences; the existing imbalanced classification methods ignore class-guided global inter-class correlations among samples, resulting in poor classification results. In order to overcome these challenges, a cost-sensitive-based TSK method (CITSK) is proposed for imbalanced epilepsy EEG signal recognition. This proposed method constructs a Class-guided Sample Collaborative Relationship (SCR) matrix to assign higher weight to positive minority samples, and supplements global inter-class collaborative penalty information on the basis of the traditional static cost-sensitive matrix. Then, a model combining the SCR matrix and TSK is proposed, which inherits the good generalization ability and interpretability of TSK while ensuring classification performance. Experiments on real EEG datasets demonstrate the effectiveness of the proposed method and its applicability to the imbalanced classification of epileptic EEG signals. Quantitatively, the CITSK method achieves an average Accuracy of 93.96%, an average F_measure of 88.33%, and an average G_means of 91.34% across various imbalanced datasets. Furthermore, comparative evaluations demonstrate that the proposed CITSK achieves superior overall performance across comprehensive evaluation metrics compared to baseline models, offering a highly robust and interpretable solution for imbalanced data scenarios via reasonable global inter-class penalty compensation.
12 August 2026 in Journal of Mechanics in Medicine and Biology
Predicting potential associations between microRNAs (miRNAs) and diseases is essential for deciphering complex pathogenic mechanisms and advancing personalized medicine. However, many existing computational methods fail to adequately capture high-order topological structures within heterogeneous biological networks, limiting their predictive performance. In this study, we propose GCN-XGB, a novel hybrid computational framework that integrates a two-layer Graph Convolutional Network (GCN) with Extreme Gradient Boosting (XGBoost) to improve the accuracy of miRNA–disease association prediction. Specifically, we first input k-mer derived initial features of miRNAs, BioBERT-derived initial features of diseases, and the computed miRNA–miRNA similarity, disease similarity, and miRNA–disease association information into the two-layer GCN. Through deep feature propagation and aggregation of second-order neighbor information, the GCN integrates these inputs to generate enhanced discriminative high-dimensional embeddings, providing more informative and discriminative features for the downstream XGBoost classifier. Experimental results under rigorous 10-fold cross-validation demonstrate that GCN-XGB consistently outperforms several state-of-the-art baseline models when all models are trained on the same GCN-enhanced features, achieving a superior AUC of 0.9601 and AUPR of 0.9587. Furthermore, case studies on prevalent neoplasms (e.g., Breast, Colon, and Lung Cancer) confirm the framework’s efficacy in discovering novel associations even in the absence of prior clinical data. Our findings suggest that GCN-XGB is a powerful and reliable tool for identifying potential disease-related miRNAs and prioritizing candidates for experimental validation.