Traditional Statistical Methods Research Articles

Prediction of compressive strength of high-volume fly ash self-compacting concrete with silica fume using machine learning techniques

The quality and composition of the components in Self-Compacting Concrete (SCC) determine its compressive strength; however, determining these complex relationships through traditional statistical methods is difficult. This study uses five state-of-the-art machine learning techniques, namely, Random Forest (RF), Adaboost, Convolutional Neural Network (CNN), K-Nearest Neighbor (KNN), and Bidirectional Long Short-Term Memory (BI-LSTM), to simulate the strength properties of high-volume fly ash self-compacting concrete (HVFA-SCC) with silica fume. A database with 240 SCC compressive strength tests that included a range of material proportions was put together in order to train these models. The primary constituents that were selected and taken into consideration as input variables were cement, fly ash, super-plasticizer, silica fume, coarse and fine aggregate, and age. Various statistical metrics, regression error characteristics curve, SHAP analysis, and uncertainty analysis were used to validate the models' effectiveness in predicting the compressive strength of HVFA-SCC with varying material compositions. The recommended RF model demonstrated the highest predictive accuracy of all the models that were analysed. While machine learning can predict compressive strengths, its opaque 'black-box' nature limits its use for SCC mix engineers. This study introduces an open-source Random Forest-based GUI to close this gap. This interface helps engineers make mix proportion decisions by accurately estimating SCC compressive strength under various test conditions.

Quantum model regression for generating fuzzy numbers in adiabatic quantum computing

This paper addresses the problem of inherent fuzziness in real-world data, arising from uncertainties, complexities, and limitations of traditional statistical methods. We introduce a pioneering method leveraging Adiabatic Quantum Computing (AQC), based on an adiabatic quantum regression model, to generate fuzzy numbers—ideal tools owing to their extension of real numbers and robust arithmetic properties. This innovative approach overcomes challenges in Quantum Machine Learning (QML) and limitations of Noisy Intermediate-Scale Quantum (NISQ) computers, providing a solution superior to conventional statistical methods and addressing the crucial issue of exponential power increase. Unique to this work, we offer rare closed-form expressions to produce fuzzy numbers through AQC and emphasise the integration of random variables and fuzzy numbers to encapsulate uncertainty fully. We propose a novel transformation of the Cumulative Distribution Function (CDF) into triangular and trapezoidal fuzzy numbers using AQC, enabling a comprehensive description of probability distributions of random variables. Experimental results are detailed, demonstrating the significant applicability and breakthroughs of this research in addressing data fuzziness.

Explainable Boosting Machine approach identifies risk factors for acute renal failure

BackgroundRisk stratification and outcome prediction are crucial for intensive care resource planning. In addressing the large data sets of intensive care unit (ICU) patients, we employed the Explainable Boosting Machine (EBM), a novel machine learning model, to identify determinants of acute kidney injury (AKI) in these patients. AKI significantly impacts outcomes in the critically ill.MethodsAn analysis of 3572 ICU patients was conducted. Variables such as average central venous pressure (CVP), mean arterial pressure (MAP), age, gender, and comorbidities were examined. This analysis combined traditional statistical methods with the EBM to gain a detailed understanding of AKI risk factors.ResultsOur analysis revealed chronic kidney disease, heart failure, arrhythmias, liver disease, and anemia as significant comorbidities influencing AKI risk, with liver disease and anemia being particularly impactful. Surgical factors were also key; lower GI surgery heightened AKI risk, while neurosurgery was associated with a reduced risk. EBM identified four crucial variables affecting AKI prediction: anemia, liver disease, and average CVP increased AKI risk, whereas neurosurgery decreased it. Age was a progressive risk factor, with risk escalating after the age of 50 years. Hemodynamic instability, marked by a MAP below 65 mmHg, was strongly linked to AKI, showcasing a threshold effect at 60 mmHg. Intriguingly, average CVP was a significant predictor, with a critical threshold at 10.7 mmHg.ConclusionUsing an Explainable Boosting Machine enhance the precision in AKI risk factors in ICU patients, providing a more nuanced understanding of known AKI risks. This approach allows for refined predictive modeling of AKI, effectively overcoming the limitations of traditional statistical models.

Application of artificial neural networks to classify Avena fatua and Avena sterilis based on seed traits: insights from European Avena populations primarily from the Balkan Region

BackgroundAvena fatua and A. sterilis are challenging to distinguish due to their strong similarities. However, Artificial Neural Networks (ANN) can effectively extract patterns and identify these species. We measured seed traits of Avena species from 122 locations across the Balkans and from some populations from southern, western, and central Europe (total over 22 000 seeds). The inputs for the ANN model included seed mass, size, color, hairiness, and placement of the awn attachment on the lemma.ResultsThe ANN model achieved high classification accuracy for A. fatua and A. sterilis (R2 > 0.99, RASE < 0.0003) with no misclassification. Incorporating geographic coordinates as inputs also resulted in successful classification (R2 > 0.99, RASE < 0.000001) with no misclassification. This highlights the significant influence of geographic coordinates on the occurrence of Avena species. The models revealed hidden relationships between morphological traits that are not easily detectable through traditional statistical methods. For example, seed color can be partially predicted by other seed traits combined with geographic coordinates. When comparing the two species, A. fatua predominantly had the lemma attachment point in the upper half, while A. sterilis had it in the lower half. A. sterilis exhibited slightly longer seeds and hairs than A. fatua, while seed hairiness and mass were similar in both species. A. fatua populations primarily had brown, light brown, and black colors, while A. sterilis populations had black, brown, and yellow colors.ConclusionsDistinguishing A. fatua from A. sterilis based solely on individual characteristics is challenging due to their shared traits and considerable variability of traits within each species. However, it is possible to classify these species by combining multiple seed traits. This approach also has significant potential for exploring relationships among different traits that are typically difficult to assess using conventional methods.

Spatio-Temporal Fluctuation Analysis of Ecosystem Service Values in Northeast China over Long Time Series: Based on Bayesian Hierarchical Modeling

Ecosystems are undergoing continuous degradation due to the dual perturbation of global climate change and human activities, posing unprecedented threats and challenges to the ecosystem services they provide. To gain a deeper understanding of the spatio-temporal evolution of ecosystem service value (ESV), it is essential to accurately capture the characteristics of its spatial and temporal changes and its influencing factors. However, traditional spatio-temporal statistical methods are limited to analyzing the heterogeneity of ESV in a single temporal or spatial dimension, which fails to meet the comprehensive analysis needs for spatio-temporal heterogeneity over an extended continuum. Therefore, this paper constructs a Bayesian spatio-temporal hierarchical model to analyze the ESV heterogeneity in both temporal and spatial dimensions in Northeast China from 2000 to 2020 to accurately identify the regions with unstable fluctuations in ESV and analyze the influencing factors behind them. It aims to comprehensively and systematically reveal the intrinsic laws of spatio-temporal evolution of ESV, and provide a scientific basis for relevant decision-making. The study found a continuous fluctuating downward trend of ESV in Northeast China from 2000 to 2020, with significant spatial and temporal heterogeneity. Notably, the distribution of hot and cold spots is regularly concentrated, especially in the transition zone from low hills to plains, which forms an “unstable zone” of spatial and temporal fluctuations of ESV. Natural factors such as NDVI and NPP exhibit a significant positive correlation with ESV, while social factors like population density and GDP show a strong negative correlation. Compared to traditional statistical methods, the Bayesian spatio-temporal hierarchical model, with its outstanding flexibility and accuracy, provides a new perspective and way of thinking for analyzing classical spatio-temporal problems. Firstly, the model examines time and space as a whole and fully accounts for the influence of spatio-temporal interactions on ESV. Secondly, the Bayesian spatio-temporal hierarchical model meets the needs of long-term continuous ESV outcome detection, which provides us with solid support for a deeper understanding of the evolution of ESV.

Exploratory analysis of new craniometric measures for the investigation of biological sex using open-access statistical and machine-learning tools on a cone-beam computed tomography sample.

Investigation of the biological sex of human remains is a crucial aspect of physical anthropology. However, due to varying states of skeletal preservation, multiple approaches and structures of interest need to be explored. This research aims to investigate the potential use of distances between bifrontal breadth (FMB), infraorbital foramina distance (IOD), nasal breadth (NLB), inter-canine width (ICD), and distance between mental foramina (MFD) for combined sex prediction through traditional statistical methods and through open-access machine-learning tools. Ethical approval was obtained from the ethics committee, and out of 100 cone beam computed tomography (CBCT) scans, 54 individuals were selected with all the points visible. Ten extra exams were chosen to test the predictors developed from the learning sample. Descriptive analysis of measurements, standard deviation, and standard error were obtained. T-student and Mann-Whitney tests were utilized to assess the sex differences within the variables. A logistic regression equation was developed and tested for the investigation of the biological sex as well as decision trees, random forest, and artificial neural networks machine-learning models. The results indicate a strong correlation between the measurements and the sex of individuals. When combined, the measurements were able to predict sex using a regression formula or machine learning based models which can be exported and added to software or webpages. Considering the methods, the estimations showed an accuracy rate superior to 80% for males and 82% for females. All skulls in the test sample were accurately predicted by both statistical and machine-learning models. This exploratory study successfully established a correlation between facial measurements and the sex of individuals, validating the prediction potential of machine learning, augmenting the investigative tools available to experts with a high differentiation potential.

Pathways to chronic disease detection and prediction: Mapping the potential of machine learning to the pathophysiological processes while navigating ethical challenges

AbstractChronic diseases such as heart disease, cancer, and diabetes are leading drivers of mortality worldwide, underscoring the need for improved efforts around early detection and prediction. The pathophysiology and management of chronic diseases have benefitted from emerging fields in molecular biology like genomics, transcriptomics, proteomics, glycomics, and lipidomics. The complex biomarker and mechanistic data from these “omics” studies present analytical and interpretive challenges, especially for traditional statistical methods. Machine learning (ML) techniques offer considerable promise in unlocking new pathways for data‐driven chronic disease risk assessment and prognosis. This review provides a comprehensive overview of state‐of‐the‐art applications of ML algorithms for chronic disease detection and prediction across datasets, including medical imaging, genomics, wearables, and electronic health records. Specifically, we review and synthesize key studies leveraging major ML approaches ranging from traditional techniques such as logistic regression and random forests to modern deep learning neural network architectures. We consolidate existing literature to date around ML for chronic disease prediction to synthesize major trends and trajectories that may inform both future research and clinical translation efforts in this growing field. While highlighting the critical innovations and successes emerging in this space, we identify the key challenges and limitations that remain to be addressed. Finally, we discuss pathways forward toward scalable, equitable, and clinically implementable ML solutions for transforming chronic disease screening and prevention.

The Role of Machine Learning in Enhancing Risk Management Strategies in Financial Institutions

Purpose: The aim of the study was to examine the role of machine learning in enhancing risk management strategies in financial institutions. Methodology: This study adopted a desk methodology. A desk study research design is commonly known as secondary data collection. This is basically collecting data from existing resources preferably because of its low cost advantage as compared to a field research. Our current study looked into already published studies and reports as the data was easily accessed through online journals and libraries. Findings: The study revealed that integration of machine learning into risk management strategies within financial institutions has demonstrated significant potential for enhancing decision-making processes and mitigating various risks. The study have consistently shown that machine learning algorithms outperform traditional statistical methods in areas such as credit risk assessment, fraud detection, market risk management, and loan portfolio optimization. These advancements have led to improved accuracy, efficiency, and timeliness in risk assessment, enabling financial institutions to make more informed decisions while reducing losses and enhancing overall performance. Unique Contribution to Theory, Practice and Policy: Modern Portfolio Theory (MPT), Efficient Market Hypothesis (EMH) &amp; Agency Theory may be used to anchor future studies on role of machine learning in enhancing risk management strategies in financial institutions. Invest in building robust data infrastructure and governance frameworks to support the implementation of machine learning models in risk management practices. High-quality data is crucial for training accurate and reliable machine learning algorithms. Establish regulatory guidelines and standards for the responsible use of machine learning in risk management within the financial industry. These guidelines should address issues such as model transparency, fairness, and accountability to ensure ethical and responsible practices.

How long-term PM exposure may affect all-site cancer mortality: Evidence from a large cohort in southern China

BackgroundEvidence of a potential causal link between long-term exposure to particulate matter (PM) and all-site cancer mortality from large population cohorts remained limited and suffered from residual confounding issues with traditional statistical methods. AimsWe aimed to examine the potential causal relationship between long-term PM exposure and all-site cancer mortality in South China using causal inference methods. MethodsWe used a cohort in southern China that recruited 580,757 participants from 2009 through 2015 and tracked until 2020. Annual averages of PM1, PM2.5, and PM10 concentrations were generated with validated spatiotemporal models. We employed a causal inference approach, the Marginal Structural Cox model, based on observational data to evaluate the association between long-term exposure to PM and all-site cancer mortality. ResultsWith an increase of 1 µg/m³ in PM1, PM2.5, and PM10, the hazard ratios (HRs) and 95% confidence interval (CI) for all-site cancer were 1.033 (95% CI: 1.025–1.041), 1.032 (95% CI: 1.027–1.038), and 1.020 (95% CI: 1.016–1.025), respectively. The HRs (95% CI) for digestive system and respiratory system cancer mortality associated with each 1 µg/m³ increase in PM1 were 1.022 (1.009–1.035) and 1.053 (1.038–1.068), respectively. In addition, inactive participants, who never smoked, or who lived in areas of low surrounding greenness were more susceptible to the effects of PM exposure, the HRs (95% CI) for all-site cancer mortality were 1.042 (1.031–1.053), 1.041 (1.032–1.050), and 1.0473 (1.025–1.070) for every 1 µg/m³ increase in PM1, respectively. The effect of PM1 tended to be more pronounced in the low-exposure group than in the general population, and multiple sensitivity analyses confirmed the robustness of the results. ConclusionThis study provided evidence that long-term exposure to PM may elevate the risk of all-site cancer mortality, emphasizing the potential health benefits of improving air quality for cancer prevention.

A Review of Time-Series Forecasting Algorithms for Industrial Manufacturing Systems

Time-series forecasting is crucial in the efficient operation and decision-making processes of various industrial systems. Accurately predicting future trends is essential for optimizing resources, production scheduling, and overall system performance. This comprehensive review examines time-series forecasting models and their applications across diverse industries. We discuss the fundamental principles, strengths, and weaknesses of traditional statistical methods such as Autoregressive Integrated Moving Average (ARIMA) and Exponential Smoothing (ES), which are widely used due to their simplicity and interpretability. However, these models often struggle with the complex, non-linear, and high-dimensional data commonly found in industrial systems. To address these challenges, we explore Machine Learning techniques, including Support Vector Machine (SVM) and Artificial Neural Network (ANN). These models offer more flexibility and adaptability, often outperforming traditional statistical methods. Furthermore, we investigate the potential of hybrid models, which combine the strengths of different methods to achieve improved prediction performance. These hybrid models result in more accurate and robust forecasts. Finally, we discuss the potential of newly developed generative models such as Generative Adversarial Network (GAN) for time-series forecasting. This review emphasizes the importance of carefully selecting the appropriate model based on specific industry requirements, data characteristics, and forecasting objectives.

Exploring Nvidia's evolution, innovations, and future stock trends

This paper undertakes a thorough examination of Nvidia's stock market performance, intertwining historical analysis with forward-looking projections to illuminate the dynamic trajectory of this semiconductor industry giant. Commencing with a retrospective review, the authors delve into pivotal milestones, technological innovations, and strategic maneuvers that have shaped Nvidia's stock evolution. Utilizing advanced machine learning algorithms, including Random Forest and Support Vector Regression (SVR), alongside traditional statistical forecasting methods, we forecast future patterns. Through the incorporation of emerging industry dynamics, technological advancements, and market forecasts, our goal is to furnish stakeholders with insights pivotal for strategic decision-making. This dual perspective, encompassing both historical retrospection and future outlook, weaves a holistic narrative capturing the essence of Nvidia's stock market journey. It serves as a valuable resource for avid readers navigating the ever-changing landscape of the semiconductor market, especially considering the expanding role of Nvidia's GPUs in Artificial Intelligence (AI) and deep learning applications.

Anomaly Detection Using Machine Learning in Hydrochemical Data From Hot Springs: Implications for Earthquake Prediction

AbstractThis study explores the potential of machine learning algorithms for earthquake prediction, utilizing fluid chemical anomaly data from hot springs. Six hot springs, located within an active fault zone along the southeastern coast of China, were carefully chosen as hydrochemical monitoring sites for an extended period of two and a half years. Using this data, a prediction model integrating six algorithms was developed to forecast M ≥ 5 earthquakes in Taiwan. The model's performance was validated against recorded earthquake events, and the factors influencing its predictive capability were analyzed. Our comprehensive analysis conclusively demonstrates the superiority of machine learning algorithms over traditional statistical methods for earthquake prediction. Additionally, including sampling time in the data sets significantly improves the model's predictive performance. However, it is important to note that the model's predictive performance varies across different hot spring and indicators type, highlighting the importance of identifying optimal indicators for specific scenarios. The model parameters, including the anomaly detection rate (P) and earthquake response time threshold (M), significantly impact the model's predictive capabilities. Therefore, adjustments are needed to optimize the model's performance for practical use. Despite limitations such as the inability to differentiate pre‐earthquake anomalies from post‐earthquake anomalies and pinpoint the precise location of earthquakes, this study successfully showcases the potential of machine learning algorithms in earthquake prediction, paving the way for further research and improved prediction methods.

Identification of Geochemical Anomalies Based on RPCA and Multifractal Theory: A Case Study of the Sidaowanzi Area, Chifeng, Inner Mongolia.

The focus of exploration geochemistry is an accurate interpretation of geochemical data and the precise extraction of anomaly information related to mineralization from complex geological information. However, geochemical data are component data and exhibit a closure effect. Thus, traditional statistical methods cannot adequately reveal and identify the distribution of deep-seated anomaly information. This paper focuses on the Sidaowanzi area in Inner Mongolia and uses multivariate component data analysis methods to process 1:50 000 soil geochemical data. Using the Exploratory Data Analysis (EDA) method, the spatial distribution and internal structure characteristics of raw, logarithmic, and isometric logarithmic ratio (ILR) transformed data were compared and, coupled with robust principal component analysis (RPCA) and elemental component biplots, the association between element combinations and mineralization indicated by these three types of data was revealed. The S-A method was used to decompose composite anomalies of the ILR transformed RPCA score data to extract the characteristics of elemental combination anomalies and background distribution, and the Fry analysis method was utilized to analyze the dominant mineralization direction within the area. The results show that (1) data transformed using the ILR eliminated the influence of the closure effect, making the data more uniform on a spatial scale and exhibiting characteristics of an approximately normal distribution. (2) The S-A method was further used to decompose the composite anomaly of the PC1 and PC2 principal component combinations. The screened-out anomaly and background fields can essentially reflect the ore-causing anomalies dominated by Au and Cu-Mo mineralization. Moreover, the extracted anomalies and background information closely align with known mineral deposits (prospects) and can effectively identify weakly retarded geochemical anomaly information. (3) Fry analysis based on geochemical anomalies indicates that the dominant mineralization directions, by an assemblage dominated by Au and Cu-Mo, predominantly occur in the NE, NW, and proximate EW orientations. The combined application of the aforementioned three methods for the quantitative analysis of geochemical data aims to explore a transferable methodological system, providing new insights and approaches for further prediction of mineralization potential.

Using isometric log-ratio in compositional data analysis for developing a groundwater pollution index

This study introduces a novel groundwater pollution index (GPI) formulated through compositional data analysis (CoDa) and robust principal component analysis (RPCA) to enhance groundwater quality assessment. Using groundwater quality monitoring data from sites impacted by the 2010–2011 foot-and-mouth disease outbreak in South Korea, CoDa uncovers critical hydrochemical differences between leachate-influenced and background groundwater. The GPI was developed by selecting key subcompositional parts (NH4+-N, Cl-, and NO3--N) using RPCA, performing the isometric log-ratio (ILR) transformation, and normalizing the results to environmental standards, thereby providing a more precise and accurate assessment of pollution. Validated against government criteria, the GPI has shown its potential as an alternative assessment tool, with its reliability confirmed by receiver operating characteristic curve analysis. This study highlights the essential role of CoDa, especially the ILR -transformation, in overcoming the limitations of traditional statistical methods that often neglect the relative nature of hydrochemical data. Our results emphasize the utility of the GPI in significantly advancing groundwater quality monitoring and management by addressing a methodological gap in the quantitative assessment of groundwater pollution.

Pancreatitis with use of new diabetic medications: a real-world data study using the post-marketing FDA adverse event reporting system (FAERS) database.

Background: Pancreatitis is characterized by inflammation of the pancreas and significantly affects quality of life. Less than 5% of pancreatitis cases are drug-induced, but recent evidence suggests a substantial risk associated with glucagon-like peptide-1 receptor agonists (GLP-1 RAs). The aim of this study was to compare the risk of developing pancreatitis between those using GLP-1 RAs and those using sodium-glucose transport protein 2 (SGLT2) inhibitors and dipeptidyl peptidase 4 (DPP-4) inhibitors. Methods: This study was done using the FDA Adverse Event Reporting System (FAERS) database from 2019 to 2021. This database contains information from diverse submissions from healthcare providers, patients, and manufacturers. To ensure fairness and accuracy, the risk of pancreatitis associated with other hypoglycemic agents (SGLT2 inhibitors and DPP-4 inhibitors) was also investigated. Traditional and Bayesian statistical analysis methods were used to identify disproportionate statistics and included the reporting odds ratio (ROR), proportional reporting ratio (PRR), empirical Bayes geometric mean (EBGM), and information component (IC). A drug-adverse-event combination that met the criteria of all four indices was deemed a signal. Results: The analysis of 2,313 pancreatitis reports linked to hypoglycemic agents revealed a predominant association with GLP-1 RA (70.2%) compared to DPP-4 inhibitors (15%) and SGLT2 (14.7%). Most of these reports involved female patients (50.4%), and the highest incidence occurred in those over 50 years old (38.4%). Additionally, 17.7% of the reports were associated with serious events. The ROR was significant for the risk of pancreatitis when using DPP-4 (13.2, 95% confidence interval (CI) 11.84-14.70), while the ROR for GLP-1 was 9.65 (95% CI 9.17-10.16). The EBGM was highest with DPP-4 (12.25), followed by GLP-1 (8.64), while IC was highest with DPP-4 inhibitors (3.61). Liraglutide had the greatest association with pancreatitis among the GLP-1 RAs (ROR: 6.83, 95% CI 6.60-7.07). Conclusion: The findings show that pancreatitis has a strong link with DPP-4 inhibitors and GPL1 agonists, which pose a greater risk. Among the GLP-1 agonist medications, liraglutide has been found to have an association with pancreatitis.

Comparison of traditional regression modeling vs. AI modeling for the prediction of dental caries: a secondary data analysis

IntroductionThere are substantial gaps in our understanding of dental caries in primary and permanent dentition and various predictors using newer modeling methods such as Machine Learning (ML) algorithms and Artificial Intelligence (AI). The objective of this study is to compare the accuracy, precision, and differences between the caries predictive capability of AI vs. traditional multivariable regression techniques.MethodsThe study was conducted using secondary data stored in the Temple University Kornberg School of Dentistry electronic health records system (axiUm) of pediatric patients aged 6–16 years who were patients on record at the Pediatric Dentistry Clinic. The outcome variables considered in the study were the decayed–missing–filled teeth (DMFT) and the decayed–extracted–filled teeth (deft) scores. The predictors included age, sex, insurance, fluoride exposure, having a dental home, consumption of sugary meals, family caries experience, having special needs, visible plaque, medications reducing salivary flow, and overall assessment questions.ResultsThe average DMFT score was 0.85 ± 2.15, while the average deft scores were 0.81 ± 2.15. For childhood dental caries, XGBoost was the best performing ML algorithm with accuracy, sensitivity. and Kappa as 81%, 84%, and 61%, respectively, followed by Support Vector Machine and Lasso Regression algorithms, both with 84% specificity. The most important variables for prediction found were age and visible plaque.ConclusionsThe machine learning model outperformed the traditional statistical model in the prediction of childhood dental caries. Data from a more diverse population will help improve the quality of caries prediction for permanent dentition where the traditional statistical method outperformed the machine learning model.

Predicting the compressive strength of carbon-enhanced cementitious composites using two-dimensional convolutional neural networks

This study addresses the limitations of traditional statistical methods and artificial neural networks (ANNs) in extracting relevant features from carbon-based cementitious composites to assess their dynamic behaviors under various environmental conditions. However, previous deep-learning methods were limited in making efficient predictions utilizing multiple factors. These shortcomings are not appropriate for predicting the properties of cement composed of various elements. In this study, these elements were converted into image form to handle efficiently the multiple factors that consist of cement, and the prediction was performed using a convolutional neural network (CNN)-based model. The concentration range of multiwalled carbon nanotubes (MWCNTs) used was 0.25 wt % to 1.0 wt %. Additionally, the specimens utilized were 50×50×50mm3 in size, and a total of 532 pieces of data were employed. Based on mean absolute error (MAE), performance was improved by approximately 25 % with ANN (MAE = 3.89) and 7.2 % with random forest (MAE = 3.14) compared with the proposed model (MAE = 2.92). The findings of this study are a novel and practical approach for predicting the mechanical properties of cementitious composites, enabling reverse application to optimize mixing ratios for CNT cement composites, and streamlining fabrication.

Machine Learning-Based Prediction of Suicidal Thinking in Adolescents by Derivation and Validation in 3 Independent Worldwide Cohorts: Algorithm Development and Validation Study.

Suicide is the second-leading cause of death among adolescents and is associated with clusters of suicides. Despite numerous studies on this preventable cause of death, the focus has primarily been on single nations and traditional statistical methods. This study aims to develop a predictive model for adolescent suicidal thinking using multinational data sets and machine learning (ML). We used data from the Korea Youth Risk Behavior Web-based Survey with 566,875 adolescents aged between 13 and 18 years and conducted external validation using the Youth Risk Behavior Survey with 103,874 adolescents and Norway's University National General Survey with 19,574 adolescents. Several tree-based ML models were developed, and feature importance and Shapley additive explanations values were analyzed to identify risk factors for adolescent suicidal thinking. When trained on the Korea Youth Risk Behavior Web-based Survey data from South Korea with a 95% CI, the XGBoost model reported an area under the receiver operating characteristic (AUROC) curve of 90.06% (95% CI 89.97-90.16), displaying superior performance compared to other models. For external validation using the Youth Risk Behavior Survey data from the United States and the University National General Survey from Norway, the XGBoost model achieved AUROCs of 83.09% and 81.27%, respectively. Across all data sets, XGBoost consistently outperformed the other models with the highest AUROC score, and was selected as the optimal model. In terms of predictors of suicidal thinking, feelings of sadness and despair were the most influential, accounting for 57.4% of the impact, followed by stress status at 19.8%. This was followed by age (5.7%), household income (4%), academic achievement (3.4%), sex (2.1%), and others, which contributed less than 2% each. This study used ML by integrating diverse data sets from 3 countries to address adolescent suicide. The findings highlight the important role of emotional health indicators in predicting suicidal thinking among adolescents. Specifically, sadness and despair were identified as the most significant predictors, followed by stressful conditions and age. These findings emphasize the critical need for early diagnosis and prevention of mental health issues during adolescence.

Online updating mode learning for streaming datasets

In the case of limited computer memory, it is worth discussing how to regression the big data stream and solve the outlier problem reasonably. To solve the above problems, this article proposes an online updatable modal linear regression algorithm, which is robust to heavy-tailed distribution, heterogeneous errors and outliers. However, due to streaming data, the traditional static statistical methods are facing new challenges. In addition, existing inference tools in online learning cannot be used directly for modal linear regression. A major challenge is that the data is one-pass and the final amount of data is not even known, which results in the inability to obtain optimal bandwidth, resulting in poor theoretical performance of the estimator. In particular, the estimation and inference methods are proposed. The process only passes the data once. Finally, the finite sample performance of the proposed method is verified by simulation and real data analysis.

Deep Time Series Forecasting Models: A Comprehensive Survey

Deep learning, a crucial technique for achieving artificial intelligence (AI), has been successfully applied in many fields. The gradual application of the latest architectures of deep learning in the field of time series forecasting (TSF), such as Transformers, has shown excellent performance and results compared to traditional statistical methods. These applications are widely present in academia and in our daily lives, covering many areas including forecasting electricity consumption in power systems, meteorological rainfall, traffic flow, quantitative trading, risk control in finance, sales operations and price predictions for commercial companies, and pandemic prediction in the medical field. Deep learning-based TSF tasks stand out as one of the most valuable AI scenarios for research, playing an important role in explaining complex real-world phenomena. However, deep learning models still face challenges: they need to deal with the challenge of large-scale data in the information age, achieve longer forecasting ranges, reduce excessively high computational complexity, etc. Therefore, novel methods and more effective solutions are essential. In this paper, we review the latest developments in deep learning for TSF. We begin by introducing the recent development trends in the field of TSF and then propose a new taxonomy from the perspective of deep neural network models, comprehensively covering articles published over the past five years. We also organize commonly used experimental evaluation metrics and datasets. Finally, we point out current issues with the existing solutions and suggest promising future directions in the field of deep learning combined with TSF. This paper is the most comprehensive review related to TSF in recent years and will provide a detailed index for researchers in this field and those who are just starting out.