EEG-based variable node functional network comparison for multiclass brain disease detection using stacked ensemble model

Estimating electricity load is paramount in the strategic planning and operation of power systems, ensuring the efficient and sustainable operation of contemporary electricity supply networks. This study proposes a Stacked Ensemble (SE) model to address these concerns and enhance short‐term net load forecasting (STNLF) by combining bidirectional long short‐term memory, support vector machine, and random forest (RF) models. An autoencoder is also used to optimise the input data, which extracts critical features from the data and enhances predictive accuracy. This SE model is designed to incorporate Explainable Artificial Intelligence (XAI) as an integral part, providing users detailed insights into input variable influences such as temperature, solar radiation, and wind speed, all interpretable and relevant, thus improving the transparency of the forecasting process. The proposed method is evaluated against real historical time‐series data in Austria from 2018 and 2019 regarding hourly measurements for temperature, solar energy, wind turbine generation, and actual loads. Results indicate that the SE model can capture complex data patterns, whereby XAI features actionable insights and proves a reasonable concurrent behaviour prediction with boundary‐specific realities. Remarkably, the proposed SE model surpassed traditional machine learning approaches on many performance metrics while showcasing robustness and reliability in increasingly improving STNLF accuracy.

The brain connectivity network (BCN) is considered to be an effective approach for analysing brain functionality. The EEG-based BCN considers electrodes as a node and functional similarity between EEGs from corresponding nodes as edge. The EEG dataset available for the evaluation might contain variable number of nodes. The variable number of nodes provides biased results while performing graph classification. Hence, the study proposed a strategy to mitigate the aforementioned challenge. The proposed method characterises variable node BCN with the help of network level metrics as a feature vector. The extracted metrics characterises the network as a whole and do not rely on the number of nodes. Two public datasets, with 16 electrodes and 19 electrodes EEG data, are used to test the suggested method. The classification is performed with the stacked-ensemble classification technique. Finally, the quantitative analysis of the proposed approach represents a significant performance with the 92.34% classification accuracy.

Train Operators can improve railway passengers’ service quality and traffic management by accurately predicting travel arrangements and delays. Precise prediction of train delays is vital for creating feasible scheduled timetables. The import of pruning stacked ensemble deep neural networks into train delay prediction helps improve model prediction accuracy and computational time. In this study, we propose a novel pruning stacked ensemble learning model that uses pruned multilayer perceptron (MLP) neural networks as a meta-learner and heterogeneous neural networks as sub-models to improve the prediction accuracy for passenger train arrival delays and evaluate the model performance using Amtrak, the most extensive US passenger railroad service data, to determine the enhanced model accuracy. We used stacked generalization regression neural networks for artificial neural networks and deep learning neural networks to create the stacked ensemble models. We optimized the models using OPTUNA and produced a pruning stacked ensemble model using the best neural network sub-models and an MLP meta-learner. Our methodology comprises data preprocessing, feature engineering, modeling, case studies, and evaluation phases. Our experiments demonstrate that our proposed pruning stacked ensemble multilayer perceptron Neural Network (PST-NN) model has various degrees of improvement in model evaluation indicators compared to the existing model by 35.20%, 53.40% in terms of accuracy and prediction error and outperforms the benchmark models by 85.22% in terms prediction error in Passenger train delay prediction respectively, which provides a new approach to building models efficiently in Passenger train arrival delay prediction.

Predicting total porosity and free fluid index is critical for the petroleum industry as it can aid in identifying the presence of hydrocarbons, determining the productivity of a reservoir, and evaluating the potential of a prospect. Machine learning techniques, such as the hybrid ensemble model developed in this study, can aid in achieving accurate predictions of total porosity and free fluid index by providing a robust fit for this problem. This study introduces a novel stacked ensemble algorithm, combining the strengths of two machine learning models to enhance the precision of estimating the total porosity and the free fluid index. The proposed hybrid ensemble model efficiently deals with the limitations of individual models by leveraging their diverse yet complementary characteristics, leading to more robust and reliable predictions. With the cost of running NMR logs not favourable and hence not always run, a custom hybrid ensemble model was used to establish a correlation between NMR-derived total porosity and free fluid volume and wireline logs. The hybrid ensemble developed used three ensemble models as base learners. The efficacy of three ensemble and deep learning models was evaluated in comparison to the hybrid ensemble. The hybrid ensemble considerably surpassed all other models based on the RMSE, R2, and MAE. This outcome implies that the hybrid ensemble can potentially estimate total porosity and the free fluid volume in the field using wireline logs. This can help improve hydrocarbon exploration and production efficiency, ultimately leading to increased productivity and reduced costs.

A respiratory disorder that attacks COVID-19 patients requires intensive supervision of medical practitioners during the isolation period. A non-contact monitoring device will be a suitable solution for reducing the spread risk of the virus while monitoring the COVID-19 patient. This study uses Frequency-Modulated Continuous Wave (FMCW) radar and Machine Learning (ML) to obtain respiratory information and analyze respiratory signals, respectively. Multiple subjects in a room can be detected simultaneously by calculating the Angle of Arrival (AoA) of the received signal and utilizing the Multiple Input Multiple Output (MIMO) of FMCW radar. Fast Fourier Transform (FFT) and some signal processing are implemented to obtain a breathing waveform. ML helps the system to analyze the respiratory signals automatically. This paper also compares the performance of several ML algorithms such as Multinomial Logistic Regression (MLR), Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), eXtreme Gradient Boosting (XGB), Light Gradient Boosting Machine (LGBM), CatBoosting (CB) Classifier, Multilayer Perceptron (MLP), and three proposed stacked ensemble models, namely Stacked Ensemble Classifier (SEC), Boosting Tree-based Stacked Classifier (BTSC), and Neural Stacked Ensemble Model (NSEM) to obtain the best ML model. The results show that the NSEM algorithm achieves the best performance with 97.1% accuracy. In the real-time implementation, the system could simultaneously detect several objects with different breathing characteristics and classify the respiratory signals into five different classes.

IntroductionVaccination is critical for reducing childhood mortality, yet completion rates for the third dose of the pentavalent vaccine (Penta 3) in East Africa remain inadequate. This study aims to predict Penta 3 vaccination dropout using a stacking ensemble machine learning model with Demographic and Health Survey (DHS) data. The objective is to identify predictors of dropout and enhance intervention strategies.MethodsThe study utilized seven base machine learning algorithms to create a stacked ensemble model with three meta-learners: Random Forest (RF), Generalized Linear Model (GLM), and Extreme Gradient Boosting (XGBoost). The H2O package facilitated the development of base learners and the stacking of super learners. Feature selection (FS) and comparisons were performed using the LASSO and Boruta algorithms. The selected features were one-hot encoded, and ordinal encoding was applied where appropriate. Hyperparameter optimization (HPO) and comparisons were conducted using grid search and random search. Model performance was assessed using five key metrics, including accuracy and the area under the curve (AUC). SHAP (Shapley Additive Explanations) values were used to interpret the model outputs and identify influential predictors. The experimental design was employed to present the results.ResultsFour experiments were conducted to evaluate feature selection and HPO methods. All stacked ensemble models outperformed individual learners, with the XGBoost meta-learner optimized with grid search and LASSO FS achieving the highest performance: 93.9% accuracy and 99.4% AUC. While RF and GLM meta-learners were also evaluated, they were outperformed by the XGBoost meta-learner. SHAP analysis revealed key features influencing Penta 3 dropout, including the place of delivery, decision-making autonomy, the mother's level of earning, and healthcare access. Home delivery increased the risk of dropout, while postnatal care by midwives and health insurance coverage lowered dropout likelihood.Conclusion and recommendationThis study provides insights into the factors influencing Penta 3 vaccination dropout in East Africa. To reduce dropout rates, interventions should focus on enhancing maternal livelihood opportunities, improving healthcare access in rural areas, and promoting institutional deliveries.

While the use of machine learning (ML) in automated real estate valuation is growing, research on stacking ML models into ensembles remains limited. In this paper, we propose a stacked ensemble model for valuing residential properties. By applying our models to a comprehensive dataset of residential real estate transactions from Jeju Island, spanning 2012 to 2021, we demonstrate that the predictive power of ML-based models can be enhanced. Our findings indicate that the stacked ensemble model, which combines predictions using ridge regression, outperforms all individual algorithms across multiple metrics. This model not only minimizes prediction errors but also provides the most stable and consistent results, as evidenced by the lowest standard deviation in both absolute errors and absolute percentage errors. Additionally, we employed the decision tree method to analyze the conditions under which specific features yield more accurate results or less reliable outcomes. It was observed that both the size and age of an apartment significantly impact prediction performance, with smaller and older complexes exhibiting lower accuracy and higher error rates.

Botnets are used for malicious activities such as cyber-attacks, spamming, and data theft and have become a significant threat to cyber security. Despite existing approaches for cyber attack detection, botnets prove to be a particularly difficult problem that calls for more advanced detection methods. In this research, a stacking classifier is proposed based on K-nearest neighbor, support vector machine, decision tree, random forest, and multilayer perceptron, called KSDRM, for botnet detection. Logistic regression acts as the meta-learner to combine the predictions from the base classifiers into the final prediction with the aim of increasing the overall accuracy and predictive performance of the ensemble. The UNSW-NB15 dataset is used to train machine learning models and evaluate their effectiveness in detecting cyber-attacks on IoT networks. The categorical features are transformed into numerical values using label encoding. Machine learning techniques are adopted to recognize botnet attacks to enhance cyber security measures. The KSDRM model successfully captures the complex patterns and traits of botnet attacks and obtains 99.99% training accuracy. The KSDRM model also performs well during testing by achieving an accuracy of 97.94%. Based on 3, 5, 7, and 10 folds, the k-fold cross-validation results show that the proposed method’s average accuracy is 99.89%, 99.88%, 99.89%, and 99.87%, respectively. Further, the demonstration of experiments and results shows the KSDRM model is an effective method to identify botnet-based cyber attacks. The findings of this study have the potential to improve cyber security controls and strengthen networks against changing threats.

This study proposes a stacked ensemble regression model for predicting annual chickpea yield based on agro-meteorological and soil data collected from six provinces in Turkey's Central Anatolia Region between 2004 and 2024. The training was carried out using Ridge Regression, Lasso Regression, and XGBoost, and the outputs of these models were combined using a Ridge meta-regressor. Input variables included log-transformed yield values, meteorological indicators (such as wind speed, temperature, humidity, and soil moisture), annual production quantities, and soil fertility parameters (nitrogen, phosphate, potassium). The dataset was split into 80% for training and 20 % for testing. On the test data, the model achieved an <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{R}^{\mathbf{2}}$</tex> score of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{0. 9 1 3}$</tex>, a MAE of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{0. 0 9 9}$</tex>, and an RMSE of 0.164, indicating high accuracy and strong generalization capability. This study demonstrates the effectiveness of ensemble learning methods for precision agriculture and offers a scalable solution for chickpea yield prediction under semi-arid conditions.

Previous studies often binarized credit ratings, but their impact on machine learning model performance has not been thoroughly examined. Wu and Giudici (2025) address this by evaluating different binarization methods with SAFE metrics (Sustainability, Accuracy, Fairness, and Explainability metrics) Babaei et al. (2025a), which are based on Random Forest (RF), Gradient Boosting Tree (GBT), Stacked Ensemble Model (SEM) and Voting Ensemble Model (VEM). In this paper, we need to analyze various classifications and identify BB=1 and BBB=1 as the best two classifications using the method of entropy, balance scores and transition matrix. Confusion matrix, ROC curve, and prediction probability distributions help determine the best-performing classification scheme. This study validates the innovativeness and applicability of SAFE metrics, emphasizing their effectiveness in distinguishing between high-risk and low-risk categories.

Head and neck cancers (HNC) are indicated when cells grow abnormally. The incidence of HNC is on the increase owing to several factors. There is often late presentation that can result in loss of lives (mortality) especially in Africa due to paucity of specialists. These challenges prompted the development of a stacked ensemble model for diagnosis of HNC to facilitate prompt referral. The data were collected which consists of 1473 instances with 18 features. Information Gain was used for selecting important features and three supervised learning algorithms were deployed for the base learners: Decision Tree (C4.5), K-Nearest Neighbors and Naïve Bayes. The predictions of the base learners were combined and passed to meta learners: Logistic Model Tree (LMT). The result showed that Information Gain method with stacked LMTwas 95.11%. It was deduced that both Information Gain with stacked MLR produced higher accuracy that the base learners’ results. Hence, this stacked model can be used for diagnosis of HNC in healthcare systems.

The proposed work aims for effective network planning and optimization in wireless communication in mmWave urban contexts for 5G and beyond by combining characteristics RMS delay spread, angle of arrival/angle of departure (AoA/AoD angles), Tx-Rx separation distance, frequency, weather, and received power. The study examines the application of machine learning algorithms for path loss prediction and the performance was evaluated using metrics such as RMSE, MAE, MAPE, and R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>. The top-performing algorithms identified were Random Forest (RF), Support Vector Regression (SVR), and Gradient Boosting (GB). R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> value was obtained using a stacking ensemble model that combined ridge regression as the meta-model with RF, SVR, and GB as base models. The stacked ensemble model shows its ability to detect complex patterns and improve prediction accuracy with low error rates (RMSE: 4.2125, MAPE: 1.8465%).

The prediction of frictional air resistance using the inherent properties of roadways is of great significance for ventilation network computation and flow regulation in underground mines. This study proposes an improved stacked learning and error correction-based prediction model for the frictional air resistance of mine airways, called friction factor. A prediction set is established by selecting ten factors, including tunnel spatial features and support forms, with the ventilation resistance coefficient as the label. The improved stacked model consists of two layers. The first layer is the base learning module, which is composed of four components: Principal Components Analysis and Back Propagation (PCA-BP), GA-Projection Pursuit Regression (GA-PPR), Random Forest (RF), LightGBM (LGBM). The second layer is the meta-learning module, which is composed of the Ridge Regression (RR). Compared to traditional stacked models, the improved model first uses the Extreme Gradient Boosting (XG Boost) learner to evaluate the significance of input feature variables to eliminate redundancy and improve accuracy, thus enhancing prediction precision and computational efficiency. Then, the first-layer prediction results are weighted based on the errors of different prediction models in the training set using K-fold cross-validation. Box-Cox transformation is applied to the training set data from the first layer to the second layer to improve prediction normality and homogeneity. The error correction prediction model extracts the historical prediction errors from the meta-learning module and constructs an error prediction model using support vector machines (SVR), which are then combined with the meta-learning results to obtain the final prediction. The improved stacked model is compared with traditional ensemble learning models and single prediction models, and quantified using three metrics: root mean square error (RMSE), mean absolute error (MAE), and R-squared (R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ). The results demonstrate that the proposed improved model effectively enhances the prediction accuracy of the ensemble learning models, providing a new prediction method for the accurate acquisition of the friction factor of mine airways.

Reinforced concrete shear walls (RCSWs) are complicated to compute their shear capacity due to their large cross-sectional height-to-thickness ratios and the fact that they are subjected to vertical loads. Numerous factors influence RCSWs’ shear strength capacity, and the analytical models find it challenging to fully account for each factor’s impact on RCSWs’ shear-bearing capacity. Machine learning (ML) technology can deeply capture the mapping relationship between each input feature and the target value, and provide a more flexible and effective prediction method for RCSW shear-bearing capacity. To this end, a shear capacity test database containing 583 RCSW specimens was first established and characterized, and then the database was employed to train single, ensemble, and deep learning models for the shear strength of shear walls and combined with hyper-parameter tuning to enhance each model’s prediction performance, after which the prediction performance of each model was compared. Then, the ML models were contrasted with conventional techniques founded on the mechanical premise. Finally, in order to improve the prediction accuracy and reliability of the ML methods, the individually trained models were integrated into a stacking model using the stacking method, and the stacking model’s prediction performance was assessed. The results of this study show that in the single model, the test set R2 of the decision tree (DT) reaches 0.94, showing good trend-capturing ability. Among the ensemble models, Gradient Boosting (GB) performs the best and is comparable to DT in terms of RMSE and R2 and significantly outperforms other ensemble methods, such as Random Forest (RF) and Bagging. Deep Neural Networks (DNNs) show the strongest predictive ability among all models, with the lowest RMSE (263 kN) and a test R2 of 0.95, which is much better than the majority of ensemble models. The ML models show high accuracy and reliability compared to the traditional RC shear wall shear capacity models. The stacking model has an R2 of 0.98 and a CoV of 0.147 in the test set, and it is much better than other independent ML models (R2 = 0.88~0.95, CoV = 0.179~0.651).

In recent years, Internet of Things (IoT)-based technologies have advanced healthcare by facilitating the development of monitoring systems, subsequently generating an exponential amount of streaming data. This streaming data can be preprocessed and analyzed using technologies that integrate ensemble models, Explainable Artificial Intelligence (XAI), feature selection (FS) method and big data streaming processing platforms to develop predictive real-time systems. This integration adds new value to healthcare that helps organizations enhance clinical decision-making, improve patient care, and elevate the overall quality of healthcare. This article presents a real-time system for the early detection and treatment of chronic kidney disease (CKD) using a real-world simulation application. The real-time system is developed in two phases. The first phase aims to propose a stacking model, apply a genetic algorithm (GA) and Particle swarm optimization (PSO) as feature selection, and explore a stacking model with the best features with explainable artificial intelligence (XAI). The best model with the best-optimized features is used to develop the second phase. The results showed that stacking model with GA is achieved the hightest performance with 100 accuracy, 100 precision, 100 recall, and 100 F1-score. The second phase is designed based on Confluent Cloud, which offers several benefits for creating a real-time streaming system based on Apache Kafka, providing multiple APIs-the Producer API and Consumer API-for data producers and consumers, respectively. Python scripts are developed to pipeline streaming data. The first Python script to generate streaming health attributes that are pushed into a Kafka topic. A second Python script to consume health attributes from a Kafka topic and apply a stacking model to predict CKD in real-time. The results showed that the stacking model with features selected by GA recorded the best performance with 100 accuracy. The pipeline's streaming steps have validated our approach's effectiveness in real-time, leveraging Confluent Cloud and Apache Kafka.