Search Results (176)

To identify hypertension, Ballistocardiograph (BCG) signals can be primarily utilized. The BCG signal must be thoroughly understood and interpreted so that its application in the classification process could become clearer and more distinct. Various unhealthy habits such as excess consumption of alcohol and tobacco, accompanied by a lack of good diet and a sedentary lifestyle, lead to hypertension. Common symptoms of hypertension include chest pain, shortness of breath, blurred vision, mood swings, frequent urination, etc. In this work, two pragmatic models are proposed for the detection of hypertension using BCG signals and machine learning models. The first model uses K-means clustering, the maximum overlap discrete wavelet transform (MODWT) and the Empirical Wavelet Transform (EWT) techniques for feature extraction, followed by the Binary Tunicate Swarm Algorithm (BTSA) and Information Gain (IG) for feature selection, as well as two efficient hybrid classifiers such as the Hybrid AdaBoost–-Maximum Uncertainty Linear Discriminant Analysis (MULDA) classifier and the Hybrid AdaBoost–Random Forest (RF) classifier for the classification of BCG signals. The second model uses Principal Component Analysis (PCA), Kernel Principal Component Analysis (KPCA) and the Random Feature Mapping (RFM) technique for feature extraction, followed by IG and the Aquila Optimization Algorithm (AOA) for feature selection, as well as two versatile hybrid classifiers such as the Hybrid AutoRegressive Integrated Moving Average (ARIMA)–AdaBoost classifier and the Time-weighted Hybrid AdaBoost–Support Vector Machine (TW-HASVM) classifier for the classification of BCG signals. The proposed methodology was tested on a publicly available BCG dataset, and the best results were obtained when the KPCA feature extraction technique was used with the AOA feature selection technique and classified using the Hybrid ARIMA–AdaBoost classifier, reporting a good classification accuracy of 96.89%. Full article

The cement and concrete industry is one of the primary sources of anthropogenic carbon dioxide (CO₂) emissions globally, responsible for nearly 8% of total emissions, making the need for a low-carbon transition urgent. CO₂ curing provides both strength enhancement and carbon sequestration, yet the compressive strength of such concrete remains challenging to predict due to limited and strongly coupled experimental factors. This study developed an explainable Automated Machine Learning (AutoML) framework with integrated uncertainty quantification to predict the 28-day compressive strength of CO₂-cured concrete. The framework was built using 198 standardized experimental data and trained with four algorithms—Random Forest (RF), Support Vector Regression (SVR), eXtreme Gradient Boosting (XGBoost), and the transformer-based Tabular Prior-Data Fitted Network (TabPFN). To enhance model accuracy and efficiency, stratified cross-validation, hyperparameter optimization, and bootstrap-based uncertainty analysis were applied during training. The results show that TabPFN achieves the highest predictive accuracy (test R² = 0.959) and maintains a stable 95% prediction interval. SHapley Additive exPlanations (SHAP) indicates that cement content, aggregate composition, water–binder (W/B) ratio, and CO₂ curing time are the dominant factors, with an optimal W/B ratio near 0.40. Interaction analysis further reveals synergistic effects between cement content and W/B, and a strengthening coupling between curing time and CO₂ concentration at longer durations. The framework enhances predictive reliability and explainability, supporting mixture design and curing optimization for low-carbon concrete development. Full article

Network vector AutoRegressive models play a vital role in multivariate time series analysis. However, previous research in the classic Network vector AutoRegressive (NAR) model is limited to strict assumptions of linearity and time-invariance of node-specific covariates. In this study, we propose a Semiparametric NAR (SNAR) model to broaden existing research horizons by (1) extending node-specific covariates to a nonlinear framework, (2) incorporating high-dimensional time-varying covariates for a more comprehensive analysis, and (3) maintaining the interpretability of the autoregressive effects of the NAR. A deep learning-based method is presented to simultaneously estimate the nonparametric function and the parameters in SNAR. We also provide theoretical proof for the convergence rate of the nonparametric deep neural network estimator to support linear-to-nonlinear extension and show that the proposed method is capable of avoiding the curse of dimensionality. Furthermore, we prove the asymptotic normality of the parametric estimators for autoregressive effects to demonstrate the maintenance of interpretability. Experiments on various numerical simulated data show that the proposed method can avoid the curse of dimensionality; for instance, in nonlinear settings, the SNAR model reduces the prediction MSE by approximately 69% compared to the classic NAR model (decreasing from 3.44 to 1.06). Furthermore, in real-world stock return analysis, the SNAR model achieves an MSE of 0.9930, significantly outperforming the NAR baseline (MSE 1.6540) and other state-of-the-art methods. Full article

Classifying the severity of intensive-care-unit-acquired muscle atrophy (ICU-AW) is essential for early prognosis and individualized neurorehabilitation, improving functional outcomes in survivors of critical illness. This study evaluated and compared advanced machine learning (ML) algorithms for classifying neuromuscular atrophy in neurocritical patients. Clinical, biochemical, anthropometric, and morphometric data from 198 neuro-ICU patients were retrospectively analyzed. Six supervised ML models—Support Vector Machine (SVM), Multilayer Perceptron (MLP), Extreme Gradient Boosting (XGBoost), TPOT AutoML, AdaBoost, and Multinomial Logistic Regression—were trained using stratified cross-validation, synthetic oversampling, and hyperparameter optimization. Among the most outstanding models, SVM achieved the best performance (accuracy = 93%, ROC-AUC = 0.95), followed by MLP (accuracy = 82.8%, ROC-AUC = 0.93) and XGBoost (accuracy = 80%, ROC-AUC = 0.94). Stability analyses across random seeds confirmed the robustness of SVM and TPOT, with the highest median AUPRC (>0.90). Explainable AI methods (LIME and SHAP) identified BMI, serum albumin, and body surface area as the most influential variables, showing physiologically consistent patterns associated with a classification of muscle loss. Full article

Venture capital investment and hedge fund investment are two asset classes of alternative investment fund portfolios. The purpose of this study was to determine whether the digital currency named bitcoin truly adds to diversification in an alternative investment fund portfolio. Vector auto regression was used to determine any unidirectional or bidirectional relationship between variables. The DCC-GARCH test was conducted to determine any conditional correlations that impact volatility transmission over a shorter and longer duration of time between variables. The results showed that there was no unidirectional or bidirectional relationship between bitcoin and FTSE venture capital index, as well as between bitcoin and the Barclays Hedge Fund Index. The DCC model showed no volatility transmission between bitcoin and the Barclays Hedge Fund Index, whereas volatility persists between bitcoin and the FTSE Venture Capital Index, connecting risk between the financial time series with only low correlations. These findings suggest that bitcoin could be used by investors, policy makers, and hedgers for diversification in alternative investment fund portfolios. Full article

This research investigates the responsiveness of climate and low-carbon (green) stock returns to oil prices and conventional energy stock returns, focusing on both contemporaneous and causal relationships, during an oil crisis. Two methodologies are used: vector auto-regressive (VAR) for testing the causal relationship, and ordinary least squares (OLS) for investigating the contemporaneous relationship. The main empirical results suggest that green stocks have a bidirectional positive contemporaneous relationship with oil prices and energy stock returns but no significant bidirectional causal relationship. The results reveal that oil prices and energy stock returns play a larger role in contemporaneous than causal relationships with green stock returns. In addition, green stock returns seem to have a stronger positive relationship with energy stock return than oil prices. Full article

Accurate and timely wheat yield prediction is valuable globally for enhancing agricultural planning, optimizing resource use, and supporting trade strategies. Study addresses the need for precision in yield estimation by applying machine-learning (ML) regression models to high-resolution Unmanned Aerial Vehicle (UAV) multispectral (MS) and Red-Green-Blue (RGB) imagery. Research analyzes five European wheat cultivars across 400 experimental plots created by combining 20 nitrogen, phosphorus, and potassium (NPK) fertilizer treatments. Yield variations from 1.41 to 6.42 t/ha strengthen model robustness with diverse data. The ML approach is automated using PyCaret, which optimized and evaluated 25 regression models based on 65 vegetation indices and yield data, resulting in 66 feature variables across 400 observations. The dataset, split into training (70%) and testing sets (30%), was used to predict yields at three growth stages: 9 May, 20 May, and 6 June 2022. Key models achieved high accuracy, with the Support Vector Regression (SVR) model reaching R² = 0.95 on 9 May and R² = 0.91 on 6 June, and the Multi-Layer Perceptron (MLP) Regressor attaining R² = 0.94 on 20 May. The findings underscore the effectiveness of precisely measured MS indices and a rigorous experimental approach in achieving high-accuracy yield predictions. This study demonstrates how a precise experimental setup, large-scale field data, and AutoML can harness UAV and machine learning’s potential to enhance wheat yield predictions. The main limitations of this study lie in its focus on experimental fields under specific conditions; future research could explore adaptability to diverse environments and wheat varieties for broader applicability. Full article

Pakistan is one of the largest recipients of remittances globally and has substantial remittance inflow fluctuations; thus, finding the remittance–gross domestic product (GDP) ratio threshold is expedient. This study examined the macroeconomic impacts of emigrant remittances in Pakistan using a vector autoregressive estimation framework and investigated the threshold of the remittance–GDP ratio that has real effects on the economy in terms of Dutch Disease and capital accumulation. The empirical results showed that, regarding the Dutch Disease effect, a remittance–GDP ratio greater than 6% leads to a decrease in the manufacturing–services ratio, whereas as for the capital accumulation effect, a remittance–GDP ratio greater than 5% leads to a decrease in the investment–consumption ratio. These outcomes suggested that emigrants’ remittance inflows in Pakistan that exceed certain levels relative to the GDP aggravate industrialisation (Dutch Disease effect) and capital accumulation. Full article

This study explores advanced process monitoring for one-shot drilling of aeronautical stacks made of aluminium 2024 and carbon fibre-reinforced polymer (CFRP) laminates using a 4.8 mm diameter drilling tool and unsupervised machine learning techniques. An experimental campaign is conducted to collect thrust force and torque signals at a 10 kHz sampling rate during the drilling process. These signals are employed for real-time process monitoring, focusing on material change detection and anomaly identification, where anomalies are defined as holes that fail to meet predefined quality criteria. An innovative approach based on unsupervised learning is proposed to enable automatic material change identification, signal segmentation, feature extraction, and hole quality assessment. Specifically, a semi-supervised approach based on a Gaussian Mixture Model (GMM) and 1D Convolutional AutoEncoder (1D-CAE) is employed to detect deviations from normal drilling conditions. The proposed method is benchmarked against state-of-the-art supervised techniques, including logistic regression (LR) and Support Vector Machines (SVMs). Results show that these traditional models struggle with class imbalance, leading to overfitting and limited generalisation, as reflected by the F1 scores of 0.78 and 0.75 for LR and SVM, respectively. In contrast, the proposed semi-supervised approach improves anomaly detection, achieving an F1 score of 0.87 by more effectively identifying poor-quality holes. This study demonstrates the potential of deep learning-based semi-supervised methods for intelligent process monitoring, enabling adaptive control in the drilling process of hybrid stacks and detecting anomalous holes. While the proposed approach effectively handles small and imbalanced datasets, further research into the application of generative AI could enhance performance, aiming for F1 scores above 0.90, thereby supporting adaptation in real industrial environments with high performance. Full article

Machine learning has attracted much attention in the field of genomic prediction due to its powerful predictive capabilities, yet the lack of an explanatory nature in modeling decisions remains a major challenge. In this study, we propose a novel machine learning method, ExAutoGP, which aims to improve the accuracy of genomic prediction and enhance the transparency of the model by combining automated machine learning (AutoML) with SHapley Additive exPlanations (SHAP). To evaluate ExAutoGP’s effectiveness, we designed a comparative experiment consisting of a simulated dataset and two real animal datasets. For each dataset, we applied ExAutoGP and five baseline models—Genomic Best Linear Unbiased Prediction (GBLUP), BayesB, Support Vector Regression (SVR), Kernel Ridge Regression (KRR), and Random Forest (RF). All models were trained and evaluated using five repeated five-fold cross-validation, and their performance was assessed based on both predictive accuracy and computational efficiency. The results show that ExAutoGP exhibits robust and excellent prediction performance on all datasets. In addition, the SHAP method not only effectively reveals the decision-making process of ExAutoGP and enhances its interpretability, but also identifies genetic markers closely related to the traits. This study demonstrates the strong potential of AutoML in genomic prediction, while the introduction of SHAP provides actionable biological insights. The synergy of high prediction accuracy and interpretability offers new perspectives for optimizing genomic selection strategies in livestock and poultry breeding. Full article

Complex systems often face significant challenges in both efficiency and performance when making long-term degradation predictions. To address these issues, this paper proposes a predictive architecture based on multi-output sparse probabilistic model regression. An adaptive health index (HI) extraction method was also introduced, which leverages unsupervised deep learning and variational mode decomposition to effectively extract health indicators from multiple measurements of complex systems. The effectiveness of the proposed method was validated using both the C-MAPSS and FLEA datasets. The case study results demonstrate that the proposed prognostic method delivered an outstanding performance. Specifically, the feature extraction method effectively reduced the measurement noise and produced robust HIs, while the multi-output sparse probabilistic model achieved lower prediction errors and a higher accuracy. Compared to traditional single-step forward-prediction methods, the proposed approach significantly reduced the time required for long-term predictions in complex systems, thus improving support for online status monitoring. Full article

In modern society with a highly developed Internet, online public opinions on swine epidemic diseases have become one of the important influencing factors for the fluctuation of pork prices. In this paper, the Baidu AI large model, Time-Varying Parameter-Stochastic Volatility-Vector Auto Regression (TVP-VAR) and Latent Dirichlet allocation (LDA) approaches are employed to investigate the dynamic impact of online public opinion regarding live swine epidemic diseases on fluctuations in pork price. The results show that: (1) Online public attention and negative sentiment exert significant time-varying impacts on pork price fluctuations, with these impacts being most pronounced in the short term and gradually diminishing over the medium and long term. (2) During the outbreaks of swine epidemic diseases, the impulse impact of online public attention and negative sentiment on pork price fluctuations exhibits distinct stage-specific characteristics. Initially, the impact is negative and subsequently turns positive before eventually waning. (3) The online discourse surrounding swine epidemic diseases can be categorized into four topics including disease transmission, vaccine technology, industry development, and disease prevention and control. Online public attention towards these four topics associated with negative sentiments generally contributes to variations in pork prices. Based on findings, several policy recommendations are proposed, including the timely release of swine epidemic disease information, the establishment and enhancement of the online public opinion monitoring and early warning system, as well as adherence to routine prevention and control of pig epidemic diseases. Full article

This study endeavors to explore the shock-transmission mechanism between Trade Policy Uncertainty (TPU) and the volatility inherent in the Gulf Cooperation Council (GCC) Islamic stock markets by employing the novel Quantile Vector Auto Regression (QVAR) with “Extended Joint” and “Frequency” domain connectedness technique. Overall findings indicated a U-shaped pattern in the shock-transmission mechanism with the higher TPU shocks transmitted towards Islamic stock market volatility at the extreme quantiles and in the long term. The “Extended Joint” QVAR connectedness approach highlights that, in bearish and moderate-volatility conditions (τ = 0.05, 0.50), diversifying portfolios across less shock-prone equity markets like Qatar and UAE can mitigate risk exposure to TPU shocks. Specific economies receiving higher TPU shocks, like Bahrain, Kuwait, and Saudi Arabia, should implement strategic frameworks, including trade credit insurance and currency hedging, for risk reduction in trade policy shocks during the bearish and moderate-volatility conditions. Conversely, Qatar and Kuwait show the least transmission of error variance from TPU during higher-volatility conditions (τ = 0.95). Moreover, the application of the Frequency-domain QVAR technique underscores the need for short-term speculators to exercise increased vigilance during bearish and bullish volatile periods, as TPU shocks can exert a more substantial influence on the Islamic equity market volatility of Bahrain, Oman, Kuwait, and Saudi Arabia. Long-term investors may need to tailor their asset-allocation strategies by increasing allocations to more stable assets that are less susceptible to TPU shocks, such as Qatar, during bearish (τ = 0.05), moderate (τ = 0.50), and bullish (τ = 0.95) volatility. Full article

This study introduces an adaptive robust approach for optimally sizing hybrid renewable energy systems (HRESs) comprising solar panels, wind turbines, batteries, and a diesel generator. It integrates vector auto-regressive models (VAR) and neural networks (NN) into dynamic uncertainty sets (DUSs) to address temporal auto-correlations and cross-correlations among uncertain parameters like energy demand and solar and wind energy supply. These DUSs are compared to static and independent dynamic uncertainty sets based on time series (TS) from the literature. An exact iterative algorithm is developed to solve the problem effectively. A case study of a northern Ontario community evaluates the proposed framework and the solution method using real test data. Simulation reveals a 10.7% increase in capital cost on average but a 36.2% decrease in operational cost, resulting in a 16.4% total cost reduction and an 8.1% improvement in system reliability compared to the nominal model employing point estimates. Furthermore, the proposed VAR- and NN-based DUSs significantly outperform classical static and TS-based dynamic sets, underscoring the necessity of considering cross-correlations in uncertainty quantification. Full article

Biomass is the primary source of renewable energy in Poland. Its share in renewable energy production in Poland has decreased in recent years, but it still maintains a nearly 70% share. Poland has extensive forest and straw resources, such as pellets, which can be used for stable biomass production. The main objective of this research was to understand the potential of plant biomass production for energy purposes in Poland and other European Union (EU) countries in terms of sustainable development. The period of analysis covered 2000–2022. Secondary data from Statistical Poland and Eurostat were used. The primary research method was the Augmented Dickey–Fuller (ADF) test, which aimed to check the stationarity of stable biomass. Moreover, we calculated the Vector Auto-Regressive (VAR) model, which was used to develop the forecast. The indigenous production of solid biomass in 2022 decreased to 363,195 TJ, while in 2018, it was 384,914 TJ. Our prognosis confirms that biomass will increase. The prognosis based on the VAR model shows an increase from 365,395 TJ in 2023 to 379,795 (TJ) in 2032. Such countries as France, Germany, Italy, Spain, Sweden, and Finland have a bigger potential for solid biomass production from forests because of their higher area. As a result, Poland’s biomass production competitiveness is varied when compared to other EU nations; it is lower for nations with a large forest share and greater for those with a low forest cover. The two main benefits of producing solid biomass are its easy storage and carbon dioxide (CO₂) neutrality. The main advantage is that solid biomass preserves biodiversity, maintains soil fertility, and improves soil quality while lowering greenhouse gas emissions and environmental pollutants. The ability to leave added value locally and generate new jobs, particularly in troubled areas, is the largest social advantage of sustained biomass production. Full article