Ensemble Machine Learning Technique Based on Gaussian Algorithm for Stream Flow Modelling

The wind power industry has experienced a significant increase and popularity in recent times, and the latest statistics indicate that this sector is still thriving. However, one of the essential steps in developing wind energy projects is finding suitable sites for wind farms, which involves understanding the nature of wind speed, wind direction, terrain, and environmental impacts. To predict the wind energy production over the expected lifespan of a wind farm, vertical wind speed extrapolation to the hub height of the wind turbine is necessary. Therefore, this study presents a comprehensive evaluation of seven statistical approaches for vertical wind speed extrapolation, including Generalized Linear Models (GLM), Linear Regression (LR), Support Vector Machines (SVM), Generalized Additive Models (GAM), Gaussian Process Regression (GPR), Regression Tree (RT), and Ensemble Regression (ER). The accuracy of these methods is assessed using performance metrics such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Normalized RMSE (NRMSE), Normalized MSE (NMSE), Mean Bias Error (MBE), Mean Absolute Error (MAE), Mean Percentage Error (MPE), Mean Absolute Percentage Error (MAPE), Symmetric Mean Absolute Percentage Error (SMAPE), and R-squared (R2). The study concludes that, on average, GLM performs the best out of all seven statistical methods.

A stock price index measures the change in several share prices, which can describe the market and assist investors in deciding on a specific investment. Thus, foreseeing the stock price index benefits investors in creating a better investment strategy. However, forecasting the stock price index can be challenging due to its non-linearity, non-stationary and high uncertainty. Gaussian process regression (GPR) is an attractive and powerful approach for prediction, especially when the data fluctuates over time with fewer restrictions. Besides, the GPR gains advantages over other forecasting techniques as it can offer predictions with uncertainty to provide margin errors. In this study, we evaluate the use of GPR to predict the stock price of Thailand (SET). The SET data are divided into 2 datasets; the data in the year 2015 - 2020 and the data in the year 2020 due to the massive change during the COVID-19 pandemic. The prediction results from the GPR are then compared to the machine learning approaches, artificial neural network (ANN) and recurrent neural network (RNN) using evaluation scores; the root mean square error (RMSE), the mean absolute error (MAE), the mean absolute percentage error (MAPE) and the Nash-Sutcliffe efficiency (NSE). The results indicate that the GPR is superior to the ANN and RNN for both datasets as it provides a high prediction accuracy. Moreover, the results suggest that the GPR is less sensitive to the number of input lags in the model. Therefore, the GPR is more favorable for the prediction of SET than the ANN and RNN.
 HIGHLIGHTS
 
 The Gaussian process regression (GPR) was applied to predict the stock price index of Thailand (SET)
 The predictive performance of the GPR was compared to artificial neural networks (ANNs) and recurrent neural networks (RNNs)
 The results indicate that GPR outperformed the other methods as it provided a high prediction accuracy along with prediction intervals
 
 GRAPHICAL ABSTRACT

A Novel Multi-model Data-Driven Ensemble Technique for the Prediction of Retention Factor in HPLC Method Development

BackgroundThe World Health Organization predicts that by 2030, chronic obstructive pulmonary disease (COPD) will be the third leading cause of death and the seventh leading cause of morbidity worldwide. Pulmonary function tests (PFT) are the gold standard for COPD diagnosis. Since COPD is an incurable disease that takes a considerable amount of time to diagnose, even by an experienced specialist, it becomes important to provide an analysis of abnormalities in a simple manner. Although many deep learning (DL) methods based on computed tomography (CT) have been developed to identify COPD, the pathological changes of COPD based on CT are multi-dimensional and highly spatially heterogeneous, and their predictive performance still needs to be improved.ObjectiveThe purpose of this study was to develop a DL-based multimodal feature fusion model to accurately estimate PFT parameters from chest CT images and verify its performance.Materials and methodsIn this retrospective study, participants underwent chest CT examination and PFT at the Fourth Clinical Medical College of Xinjiang Medical University between January 2018 and July 2024. In this study, the 1-s forced expiratory volume (FEV1), forced vital capacity (FVC), 1-s forced expiratory volume ratio forced vital capacity (FEV1/FVC), 1-s forced expiratory volume to predicted value (FEV1%), and forced vital capacity to predicted value (FVC%) of PFT parameters were used as predictors and the corresponding chest CT of 3108 participants. The data were randomly assigned to the training group and the validation group at a ratio of 9:1, and the model was cross-validated using 10-fold cross-validation. Each parameter was trained and evaluated separately on the DL network. The mean absolute error (MAE), mean squared error (MSE), and Pearson correlation coefficient (r) were used as evaluation indices, and the consistency between the predicted and actual values was analyzed using the Bland-Altman plot. The interpretability of the model’s prediction process was analyzed using the Grad-CAM visualization technique.ResultsA total of 2408 subjects were included (average age 66 ± 12 years; 1479 males). Among these, 822 cases were used for encoder training to extract image features, and 1,586 cases were used for the development and validation of a multimodal feature fusion model based on a multilayer perceptron (MLP). The MAE, MSE, and r predicted between PFT and model estimates for FEV1 were 0.34, 0.20, and 0.84, respectively. For FVC, the MAE, MSE, and r were 0.42, 0.31, and 0.81, respectively. For FEV1/FVC, the MAE, MSE, and r were 6.64, 0.73, and 0.77, respectively. For FEV1%, the MAE, MSE, and r were 13.42, 3.01, and 0.73, respectively. For FVC%, the MAE, MSE, and r were 13.33, 2.97, and 0.61, respectively. It was observed that there was a strong correlation between the measured and predicted indices of FEV1, FVC, FEV1/FVC, and FEV1%. The Bland-Altman plot analysis showed good consistency between the estimated values and the measured values of all PFT parameters.ConclusionsThe preliminary research results indicate that the MLP-based multimodal feature fusion model has the potential to predict PFT parameters in COPD patients in real time. However, it is worth noting that the study used indicators before the use of bronchodilators, which may affect the interpretation of the results. Future studies should use measurements taken after bronchodilator administration to better align with clinical standards.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12890-025-03957-7.

The increasing penetration of renewable energy sources and the variability of demand in microgrids have intensified the need for accurate short-term load forecasting (STLF). This study develops a multivariable STLF framework using the Bayesian Regularization Algorithm (BRA) and validates its performance against a Bidirectional Long Short-Term Memory (Bi-LSTM) network. Temperature, humidity, and feeder current were employed as input variables, while electrical load (MW) served as the output. Both models were implemented in MATLAB R2018a with a 120-minute forecasting horizon. Performance evaluation was conducted using regression analysis and statistical metrics, including Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and the coefficient of determination (R²). The results show that the BRA model achieved superior regression accuracy (R² ≈ 0.997), demonstrating robustness and generalization under noisy conditions. By contrast, Bi-LSTM attained R² ≈ 0.991 with lower error magnitudes (MAE = 0.0293, RMSE = 0.0412), reflecting its strength in capturing sequential dependencies. A comparative review with existing literature, including confirms the competitiveness of Bi-LSTM while highlighting BRA’s resilience against noisy fluctuations typical of Nigerian microgrid environments. This study contributes to microgrid energy management by providing practical insights into selecting forecasting models, emphasizing that BRA is well-suited for noisy, small datasets, while Bi-LSTM excels in sequence-sensitive forecasting. The findings also suggest the potential for hybrid BRA–Bi-LSTM architectures to enhance forecasting accuracy in real-world microgrid applications.

Accurate forecasts of water demand are a crucial factor in the strategic planning and judicious use of finite water resources within a region, underpinning sustainable socio-economic development. This study aims to compare the applicability of various artificial intelligence models for long-term water demand forecasting across different water use sectors. We utilized the Tuojiang River basin in Sichuan Province as our case study, comparing the performance of five artificial intelligence models: Genetic Algorithm optimized Back Propagation Neural Network (GA-BP), Extreme Learning Machine (ELM), Gaussian Process Regression (GPR), Support Vector Regression (SVR), and Random Forest (RF). These models were employed to predict water demand in the agricultural, industrial, domestic, and ecological sectors using actual water demand data and relevant influential factors from 2005 to 2020. Model performance was evaluated based on the Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE), with the most effective model used for 2025 water demand projections for each sector within the study area. Our findings reveal that the GPR model demonstrated superior results in predicting water demand for the agricultural, domestic, and ecological sectors, attaining R2 values of 0.9811, 0.9338, and 0.9142 for the respective test sets. Also, the GA-BP model performed optimally in predicting industrial water demand, with an R2 of 0.8580. The identified optimal prediction model provides a useful tool for future long-term water demand forecasting, promoting sustainable water resource management.

Accurate ultra-short-term wind speed forecasting is great significance to ensure large scale integration of wind power into the power grid, but the randomness, instability, and non-linear nature of wind speed make it very difficult to be predicted accurately. To solve this problem, shifted window stationary attention transformer (SWSA transformer) is proposed based on a global attention mechanism for ultra-short-term forecasting of wind speed. SWSA transformer can sufficiently extract these complicated features of wind speed to improve the prediction accuracy of wind speed. First, positional embedding and temporal embedding are added at the bottom of the proposed method structure to mark wind speed series, which enables complicated global features of wind speed to be more effectively extracted by attention. Second, a shifted window is utilized to enhance the ability of attention to capture features from the edge sequences. Third, a stationary attention mechanism is applied to not only extract features of wind speed but also optimize the encoder-decoder network for smoothing wind speed sequences. Finally, the predicted values of wind speed are obtained using the calculation in the decoder network. To verify the proposed method, tests are performed utilizing data from an real offshore wind farm. The results show that the proposed method outperforms many popular models evaluated by many indexes including gated recurrent unit, Gaussian process regression, long-short term memory, shared weight long short-term memory network, and shared weight long short-term memory network -Gaussian process regression, in terms of mean absolute error, mean square error (MSE), root mean square error, mean absolute percentage error, mean square percentage error, and coefficient of determination (R2).

In this study, new learning algorithms were employed viz: grey-black-box (GBB) and kernel optimization(K-SVR) for short-term load demand forecasting. The obtained data is randomly categorized into calibration and verification phases. The prediction accuracy of the algorithm is assessed using Correlation Coefficient (R), Coefficient of Determination (R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ), Mean Square Error (MSE), and Root Mean Square Error (RMSE). The model combinations were achieved using the input variables feature selection approach. The GBB model has the highest goodness-of-fit (99%) and lowest prediction error (0.0). The results for both GBB, K-SVR models proved promising despite outstanding reliability of GBB model. The results show that the new learning algorithms are capable of forecasting load demand with simultaneous accuracy higher than 80%.

COVID-19 first appeared in China in December 2019. In Türkiye, the first case was seen on March 11, 2020. Since the emergence of COVID-19, it has rapidly affected the world and caused the death of many people. The rapid spread of the virus around the world and the fact that it poses a global threat have made it mandatory for the authorities to take quick measures and decisions. For this reason, the diagnosis, treatment, and prediction of the number of cases of COVID-19 disease is a vital issue. In this paper, daily deaths, daily cases, cumulative deaths, and cumulative case numbers were predicted using Türkiye's COVID-19 data. The prediction analysis for COVID-19 was conducted employing artificial intelligence (AI) techniques, specifically Stacked Long-Short Term Memory (Stacked-LSTM), Bidirectional Long Short-Term Memory (Bi-LSTM), and Artificial Neural Network (ANN) methods. Hyperparameter optimization using the Gray Wolf Optimizer Algorithm (GWO) was applied to optimize these models. To assess the accuracy and efficacy of prediction models, various performance metrics were employed, including Mean Absolute Percentage Error (MAPE), Coefficient of Determination (R^2 Score), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), Mean Squared Error (MSE), and Explained Variance Score (EVS). The assessment of the accuracy and efficacy of the pandemic prediction models involved comparing the predicted data values with the actual values. Based on this analysis, the prediction model that exhibited the highest degree of similarity to the actual values was determined to be the most reliable and effective.

Stock price prediction is a crucial, yet highly challenging task due to the nonlinear, volatile, and complex nature of financial markets. Traditional forecasting methods often struggle to capture intricate patterns in real-world data. This study proposes a machine learning-based approach using Gaussian Process Regression (GPR) to predict short-term stock prices, specifically targeting the Open, High, and Low values of stocks listed under the NIFTY 500 index. The model incorporates a rich set of financial indicators including previous close price, trading volume, 52-week highs/lows, and the percentage increase or decrease in value over the past 30 days and past 1 year. The dataset is cleaned, log-transformed to reduce skewness in highly varying financial attributes, scaled and evaluated using k-fold cross-validation to ensure robust performance estimation. The GPR model is trained with a composite kernel combining Radial Basis Function (RBF), WhiteKernel, and ConstantKernel. The model outputs not only point predictions but also provides uncertainty estimations, making it highly suitable for real-world financial decision-making. Evaluation metrics such as Mean Squared Error (MSE), <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathrm{R}^{2}$</tex> score, Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE) are computed, and the predictions are visualized for interpretability. Experimental results demonstrate that the proposed model is both accurate and interpretable, with the ability to support investor decisions through data-driven insights. The system achieves a MAPE of 0.748567% and an <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathrm{R}^{2}$</tex> score of 0.997453, indicating promising performance in forecasting key stock indicators.

Analysis of aggregated load consumption forecasting in short, medium and long term horizons using Dynamic Mode Decomposition

Air quality improvement is directly associated with the Sustainable Development Goals (SDGs), established by the United Nations in 2015. To reduce potential impacts from air pollution, computational intelligence by supervised machine learning, using different artificial neural networks (ANNs) techniques, shows to be a promising tool. To enhance their abilities to predict air quality, ANNs have been combined with data preprocessing. The present work performs short-term forecasting of hourly ground-level ozone using long short-term memory (LSTM), a type of recurrent neural network, with the discrete wavelet transform. The study was performed using data from a tropical coastal-urban site in Southeast Brazil, highly influenced by intense convective weather with complex terrain. The models’ performance was carried out by comparing statistical indices of errors and agreement, namely: mean squared error (MSE), normalized mean squared error (NMSE), mean absolute error (MAE), Pearson’s r, R2 and mean absolute percentage error (MAPE). When comparing the statistical metrics values, it is shown that the combination of artificial neural networks with wavelet transform enhanced the model’s ability to forecast ozone levels compared to the baseline model, which did not use wavelets.KeywordsOzone forecastingAir qualityNeural networksDeep learning

Accurate estimation of crop parameters, such as Leaf Area Index (LAI) and biomass over large areas using remote sensing techniques, is crucial for monitoring crop growth and yield prediction. In this study, a Gaussian Process Regression (GPR) method was developed to estimate LAI and biomass values of sugarcane during growth season using optical and synthetic-Aperture Radar (SAR) data fusion. Predicting LAI on an independent test data set using the GPR and the combined optical and SAR indices provided better prediction accuracies of LAI; with the GPR based on radial basis function (Root Mean Square Error [RMSE] = 0.34, Mean Absolute Error [MAE] = 0.28 and Mean Absolute Percentage Error [MAPE] = 10.5%) and polynomial function (RMSE = 0.42, MAE = 0.31 and MAPE = 12.58%), respectively. The test results of sugarcane biomass also showed that the GPR (poly) produced the highest statistical results (RMSE = 2.45 kg/m2, MAE = 1.72 kg/m2, MAPE = 8.1%) using the combined indices. The results suggest that the crop biophysical retrieval based on optical and SAR data fusion and GPR proposed in this study could improve LAI and biomass estimation that could help for effective crop growth monitoring and mapping applications.

The increasing demand for electrical energy in Indonesia encourages the availability of a reliable, efficient and stable power distribution system, especially in maintaining voltage quality during the transmission to distribution process. One of the main challenges in the distribution system is the occurrence of voltage drops which if not controlled can reduce energy efficiency and damage electrical equipment. In overcoming these problems, this research applies the Artificial Neural Network (ANN) method based on the backpropagation algorithm as an alternative approach in analyzing and predicting voltage drops in power distribution networks. Simulations were conducted using MATLAB R2024b software with input variables in the form of voltage, power factor, conductor length and diameter, cable type resistance, reactance, and load current. Meanwhile, the output variable analyzed is voltage drop. The simulation results show that the ANN model has a high predictive ability indicated by the Mean Square Error (MSE) value of 0.012743, Root Mean Square Error (RMSE) of 0.11289, Mean Absolute Error (MAE) of 0.033586, and Mean Absolute Percentage Error (MAPE) of 6.4292 %. This low error value indicates that the ANN is able to recognize complex patterns between input variables and provide adaptive responses to various variations in distribution system conditions so that the application of ANN using the backpropagation algorithm proves effective and efficient as a tool in supporting the technical decision-making process.

In this work, an artificial neural network (ANN) model was developed with the aim of predicting fouling resistance for heat exchanger, the network was designed and trained by means of 375 experimental data points that were selected from the literature. These data points contain six inputs, including time, volumetric concentration, heat flux, mass flow rate, inlet temperature, thermal conductivity and fouling resistance as an output. The experimental data are used for training, testing and validation of the ANN using multiple layer perceptron (MLP). The comparison of statistical criteria of different networks shows that the optimal structure for predicting the fouling resistance of the nanofluid is the MLP network with 20 hidden neurons, which has been trained with Levenberg-Marquardt (LM) algorithm. The accuracy of the model was assessed based on three known statistical metrics including mean square error (MSE), mean absolute percentage error (MAPE) and coefficient of determination (R2). The obtained model was found with the performance of {MSE = 6.5377 × 10-4, MAPE = 2.40% and R2 = 0.99756} for the training stage, {MSE = 3.9629 × 10-4, MAPE = 1.8922% and R2 = 0.99835} for the test stage and {MSE = 5.8303 × 10-4, MAPE = 2.57% and R2 = 0.99812} for the validation stage. In order to control the fouling procedure, and after conducting a sensitivity analysis, it found that all input variables have strong effect on the estimation of the fouling resistance.