Performance Of Individual Models Research Articles

An Analysis of the Effect of Skew Rolling Parameters on the Surface Quality of C60 Steel Parts Using Classification Models.

This paper presents the experimental and numerical results of a study on producing axisymmetric parts made of the C60-grade steel by skew rolling. The experimental part of this study involved conducting the skew rolling process with varying parameters, including the forming angle α, tool angle θ, chuck velocity Vu, and reduction ratio δ. Their effect on the quality of produced parts was examined and described by the roughness parameter Ra. Numerical calculations involved the use of machine learning models to predict the quality class of produced parts. The highest prediction accuracy of the results was obtained with the random forest and logistic regression models. Metrics such as precision, recall and accuracy were used to evaluate the performance of individual models. Confusion matrices and ROC curves were also employed to illustrate the performance of the classification models. The results of this study will make it possible to prevent the formation of spiral grooves on the circumference of steel parts during the rolling process.

Prediction of family inheritance in non-sarcomeric hypertrophic cardiomyopathy using clinical variables and 12-lead electrocardiogram artificial intelligence models

Abstract Background Hypertrophic cardiomyopathy (HCM) can be familial even in the absence of a causative sarcomeric variant, justifying periodical screening of first-degree relatives. However, indiscriminate screening may burden low-risk individuals unnecessarily. The prevalence and predictors of familial disease in non-sarcomeric HCM remain inadequately characterized. Purpose Our study aims to develop a tailored approach to family screening based on index cases’ clinical profiles. By leveraging artificial intelligence (AI) models, we aim to predict the likelihood of family disease using pertinent clinical variables and electrocardiograms (ECGs) from the proband. Methods From January 2008 to June 2022, all probands diagnosed with HCM at our Heart Institute genetics clinic were included if they had i) negative genetic testing for a (likely) pathogenic variant in a sarcomeric gene and ii) ≥1 first- or second-degree relative screened for HCM. The outcome was the presence of family disease defined as left ventricular wall thickness ≥13mm or verified autopsy report confirming HCM. Predictors of family disease were assessed at the first genetic clinic visit and considered age, sex, hypertension, and echocardiography metrics (maximal wall thickness, septal morphology, outflow tract obstruction at rest or Valsava). Using 12-lead ECGs and clinical variables, we used machine learning to train two distinct AI models. Model A utilized ECG signals only to train a ResNet model, while Model B used both ECGs and selected clinical variables (below 0.2 alpha level in logistic regressions) to train a random forest model to predict family disease. We performed evaluations at both ECG-level (one ECG used per patient) and patient-level (all available ECGs used in a patient) and we present individual and ensemble model performance. Results Among 356 families with sarcomere-negative HCM, 106 (30%) were familial. Probands were aged 53 ± 16 years, 30% female, 36% with a diagnosis of hypertension and 43% with peak outflow tract obstruction ≥30mmHg. In univariable logistic regressions, family disease was associated with age at diagnosis (OR 0.89 per 5-year, P=0.003), hypertension (OR 0.60, P=0.046) and obstruction (OR 0.52, P=0.013). The ensemble AI model, which integrated clinical and ECG data, achieved an AUC of 0.82 (95% CI: [0.795, 0.848]). It demonstrated a specificity of 0.891 (95% CI: [0.867, 0.913]) and a negative predictive value of 0.798 (95% CI: [0.772, 0.823]). When comparing ECG-level and patient-level data (1 to 10 ECGs per patient), the models consistently achieved AUC values between 0.78 and 0.81, indicating stable performance across various data subsets. Conclusion The prevalence of family inheritance in sarcomere-negative HCM can be accurately predicted using AI models that integrate clinical variables and 12-lead ECG data. This approach could enable tailored screening strategies based on the likelihood of family disease.

A multi-layer nesting and integration approach for predicting groundwater levels in agriculturally intensive areas using data-driven models

Agricultural water demand, groundwater extraction, surface water delivery and climate exhibit complex nonlinear relationships with groundwater storage in agricultural regions. As alternatives to computationally intensive physical models, data-driven machine learning methods are frequently employed as surrogates to capture these complex relationships, owing to their high computational efficiency. Inevitably, reliance on a single machine learning model may lead to underestimation of prediction uncertainty and potentially result in reduced accuracy. This paper presents a multi-layer nesting and integration approach for predicting groundwater levels in agriculturally intensive areas using data-driven models. The key contributions of the research are threefold: 1) the development of a comprehensive input variable selection process considering the lag effects and driving mechanisms of groundwater level variations; 2) the implementation of an optimization-based hyperparameter tuning method to enhance the performance of individual machine learning models; and 3) the establishment of a multi-model integration framework based on a multi-layer nesting technique. This approach combines the outputs of multiple machine learning models to consolidate predictions and expand the hypothesis space. The effectiveness of the proposed framework is demonstrated through case studies in the Huaihe River Basin, a major agricultural region in China. The results show that the multi-layer nesting and integration approach outperforms the use of individual machine learning models, providing more accurate and reliable groundwater level predictions. This framework offers valuable insights for decision-makers and water resource managers, supporting sustainable groundwater management and addressing the challenges faced in agriculturally intensive areas.

Spatio-Temporal Heterogeneous Ensemble Learning Method for Predicting Land Subsidence

The prediction of land subsidence is of significant value for the early warning and prevention of geological disasters. Although numerous land subsidence prediction methods are currently available, two obstacles still exist: (i) spatio-temporal heterogeneity of land subsidence is not well considered, and (ii) the prediction performance of individual models is unsatisfactory when the data do not meet their assumptions. To address these issues, we developed a spatio-temporal heterogeneous ensemble learning method for predicting land subsidence. Firstly, a two-stage hybrid spatio-temporal clustering method was proposed to divide the dataset into internally homogeneous spatio-temporal clusters. Secondly, within each spatio-temporal cluster, an ensemble learning strategy was employed to combine one time series prediction model and three spatio-temporal prediction models to reduce the prediction uncertainty of an individual model. Experiments on a land subsidence dataset from Cangzhou, China, show that the prediction accuracy of the proposed method is significantly higher than that of four individual prediction models.

A new understanding of the alkali-silica reaction expansion in concrete using a hybrid ensemble model

Alkali-Silica Reaction (ASR) is a critical issue affecting the durability of concrete structures, leading to significant long-term maintenance costs and safety concerns. This study addresses the need for accurate prediction of ASR-induced expansion in concrete, leveraging advanced machine learning techniques. Two hybrid ensemble methods, the inverse variance method and the artificial neural network-based ensemble method, were proposed to integrate the individual models to enhance the predictive capabilities. A comprehensive experimental database comprising over 1997 sets of ASR expansion data was collected from relevant literature with the environmental conditions, specimen geometry and material composition as the input variables and specimen strain as the output variable. The findings indicate that the predictive performance of individual models falls short of the performance of the hybrid ensemble machine learning model, with the inverse variance-based ensemble model exhibiting the highest performance (R = 97.2 %, RMSE = 0.066 mm). The Shapley Additive Explanations analysis reveals that the most influential factors include reaction days, alkali content, aggregate particle size, and silica content, contributing to up to 35 % of the variation in ASR expansion. This work offers valuable insights for the efficient assessment of durability of concrete structures concerning ASR expansion.

Exploring Thematic Diversity in Classical Chinese Poetry: A Novel Dataset and a BERT-enhanced Ensemble Learning Approach

Classical Chinese poetry, as an essential aspect of cultural heritage, exhibits rich theme diversity often overlooked in natural language processing research. To address this gap, we aim to explore the classification of thematic categories within this literary domain. We curate a dataset of 2,918 annotated poems spanning seven common themes and propose a BERT-based ensemble learning approach for effective classification. Although this method integrates existing models, it achieves an accuracy and F1 score of over 72% in the 7-class task, surpassing established baselines, and providing a baseline for future research. The experimental findings reveal the effectiveness of ensemble strategies in improving individual base model performance and highlight the potential of the MLP-based ensemble technique. The study contributes to a deeper understanding of thematic categories and textual features in classical Chinese poetry, and offers an automated classification system for classical Chinese poems.

Multimodal Affective Communication Analysis: Fusing Speech Emotion and Text Sentiment Using Machine Learning

Affective communication, encompassing verbal and non-verbal cues, is crucial for understanding human interactions. This study introduces a novel framework for enhancing emotional understanding by fusing speech emotion recognition (SER) and sentiment analysis (SA). We leverage diverse features and both classical and deep learning models, including Gaussian naive Bayes (GNB), support vector machines (SVMs), random forests (RFs), multilayer perceptron (MLP), and a 1D convolutional neural network (1D-CNN), to accurately discern and categorize emotions in speech. We further extract text sentiment from speech-to-text conversion, analyzing it using pre-trained models like bidirectional encoder representations from transformers (BERT), generative pre-trained transformer 2 (GPT-2), and logistic regression (LR). To improve individual model performance for both SER and SA, we employ an extended dynamic Bayesian mixture model (DBMM) ensemble classifier. Our most significant contribution is the development of a novel two-layered DBMM (2L-DBMM) for multimodal fusion. This model effectively integrates speech emotion and text sentiment, enabling the classification of more nuanced, second-level emotional states. Evaluating our framework on the EmoUERJ (Portuguese) and ESD (English) datasets, the extended DBMM achieves accuracy rates of 96% and 98% for SER, 85% and 95% for SA, and 96% and 98% for combined emotion classification using the 2L-DBMM, respectively. Our findings demonstrate the superior performance of the extended DBMM for individual modalities compared to individual classifiers and the 2L-DBMM for merging different modalities, highlighting the value of ensemble methods and multimodal fusion in affective communication analysis. The results underscore the potential of our approach in enhancing emotional understanding with broad applications in fields like mental health assessment, human–robot interaction, and cross-cultural communication.

Predicting Grapevine Physiological Parameters Using Hyperspectral Remote Sensing Integrated with Hybrid Convolutional Neural Network and Ensemble Stacked Regression

Grapevine rootstocks are gaining importance in viticulture as a strategy to combat abiotic challenges, as well as enhance scion physiology. Direct leaf-level physiological parameters like net assimilation rate, stomatal conductance to water vapor, quantum yield of PSII, and transpiration can illuminate the rootstock effect on scion physiology. However, these measures are time-consuming and limited to leaf-level analysis. This study used different rootstocks to investigate the potential application of aerial hyperspectral imagery in the estimation of canopy level measurements. A statistical framework was developed as an ensemble stacked regression (REGST) that aggregated five different individual machine learning algorithms: Least absolute shrinkage and selection operator (Lasso), Partial least squares regression (PLSR), Ridge regression (RR), Elastic net (ENET), and Principal component regression (PCR) to optimize high-throughput assessment of vine physiology. In addition, a Convolutional Neural Network (CNN) algorithm was integrated into an existing REGST, forming a hybrid CNN-REGST model with the aim of capturing patterns from the hyperspectral signal. Based on the findings, the performance of individual base models exhibited variable prediction accuracies. In most cases, Ridge Regression (RR) demonstrated the lowest test Root Mean Squared Error (RMSE). The ensemble stacked regression model (REGST) outperformed the individual machine learning algorithms with an increase in R2 by (0.03 to 0.1). The performances of CNN-REGST and REGST were similar in estimating the four different traits. Overall, these models were able to explain approximately 55–67% of the variation in the actual ground-truth data. This study suggests that hyperspectral features integrated with powerful AI approaches show great potential in tracing functional traits in grapevines.

Ensemble Machine Learning Classification Models for Predicting Pavement Condition

Forecasting pavement performance condition is essential within the pavement management system to optimize decisions with regard to planning maintenance and rehabilitation projects. Accurate forecasts facilitate timely interventions and assist in formulating cost-effective asset management plans. Data-driven machine learning models that utilize historical data to improve forecasting precision have gained attention in the field of asset management. Although numerous studies have employed regression-based models to forecast pavement condition, transportation asset management often operates according to condition index ranges rather than exact values. Therefore, classification models are suitable for predicting pavement condition grades and determining the appropriate maintenance type for pavement assets. This research focuses on developing five machine learning classification models to predict pavement condition: random forest; gradient boost; support vector machine; k-nearest neighbors; and artificial neural network. To enhance prediction performance, these models are integrated using ensemble methods, including voting and stacking. The classification models are developed using a dataset from the Georgia Department of Transportation that documented the condition of asphalt pavements for predefined maintenance sections between 2017 and 2021. A voting ensemble model constructed with the two best-performing individual classification models reached the highest accuracy rate at 83%. Although the performance of individual models fluctuates, ensemble models consistently produce a top-tier performance regardless of the variations in data sampling. Therefore, ensemble methods are recommended for developing pavement condition prediction models to improve accuracy and achieve a more consistent quality of predictions. The findings of this research will provide transportation agencies with information to help them strengthen their forecasting practices in relation to pavement condition, thereby improving their maintenance planning and cost savings.

MRI-based clinical-radiomics nomogram model for predicting microvascular invasion in hepatocellular carcinoma.

Preoperative microvascular invasion (MVI) of liver cancer is an effective method to reduce the recurrence rate of liver cancer. Hepatectomy with extended resection and additional adjuvant or targeted therapy can significantly improve the survival rate of MVI+ patients by eradicating micrometastasis. Preoperative prediction of MVI status is of great clinical significance for surgical decision-making and the selection of other adjuvant therapy strategies to improve the prognosis of patients. Established a radiomics machine learning model based on multimodal MRI and clinical data, and analyzed the preoperative prediction value of this model for microvascular invasion (MVI) of hepatocellular carcinoma (HCC). The preoperative liver MRI data and clinical information of 130 HCC patients who were pathologically confirmed to be pathologically confirmed were retrospectively studied. These patients were divided into MVI-positive group (MVI+) and MVI-negative group (MVI-) based on postoperative pathology. After a series of dimensionality reduction analysis, six radiomic features were finally selected. Then, linear support vector machine (linear SVM), support vector machine with rbf kernel function (rbf-SVM), logistic regression (LR), Random forest (RF) and XGBoost (XGB) algorithms were used to establish the MVI prediction model for preoperative HCC patients. Then, rbf-SVM with the best predictive performance was selected to construct the radiomics score (R-score). Finally, we combined R-score and clinical-pathology-image independent predictors to establish a combined nomogram model and corresponding individual models. The predictive performance of individual models and combined nomogram was evaluated and compared by receiver operating characteristic curve (ROC). Alpha-fetoprotein concentration, peritumor enhancement, maximum tumor diameter, smooth tumor margins, tumor growth pattern, presence of intratumor hemorrhage, and RVI were independent predictors of MVI. Compared with individual models, the final combined nomogram model (AUC: 0.968, 95% CI: 0.920-1.000) constructed by radiometry score (R-score) combined with clinicopathological parameters and apparent imaging features showed the optimal predictive performance. This multi-parameter combined nomogram model had a good performance in predicting MVI of HCC, and had certain auxiliary value for the formulation of surgical plan and evaluation of prognosis.

A novel hybrid ensemble model for mineral prospectivity prediction: A case study in the Malipo W-Sn mineral district, Yunnan Province, China

The utilization of machine learning and data-driven approaches for mineral prospectivity modeling has become increasingly important in supporting regional mineral exploration. However, supervised learning methods have encountered several common challenges, such as overfitting and underfitting, in the process of mineral prospectivity modeling, particularly due to imbalanced sample sizes used for model training. This imbalance could limit the generalization capability of models and result in inadequate predictive performance of individual models. The present study introduced a novel hybrid ensemble model (HEM) that integrates various learner methods, including generative adversarial network (GAN) and transfer learning techniques, to effectively address the issues above. These intelligent prediction techniques were applied to analyze the geological, geochemical, and geophysical dataset in order to delineate potential areas for W-Sn mineralization in Malipo County, Yunnan Province, China. Meanwhile, some well-established supervised learning methods such as Convolutional Neural Networks (CNN), Boosting-based model (e.g., LightGBM), and Bagging-based models, which demonstrated exceptional performance in dealing with classification issues were also integrated into the HEM. The weights for integrating these models were determined automatically using the equidistant-sampling technique. The superior performance of the HEM in identifying known deposits made it an efficient tool for mineral prediction mapping. Additionally, this model also offered high flexibility and scalability by allowing adjustments to the types and quantities of base learners based on data characteristics.

Fusing Diverse Decision Rules in 3D-Radiomics for Assisting Diagnosis of Lung Adenocarcinoma

This study aimed to develop an interpretable diagnostic model for subtyping of pulmonary adenocarcinoma, including minimally invasive adenocarcinoma (MIA), adenocarcinoma in situ (AIS), and invasive adenocarcinoma (IAC), by integrating 3D-radiomic features and clinical data. Data from multiple hospitals were collected, and 10 key features were selected from 1600 3D radiomic signatures and 11 radiological features. Diverse decision rules were extracted using ensemble learning methods (gradient boosting, random forest, and AdaBoost), fused, ranked, and selected via RuleFit and SHAP to construct a rule-based diagnostic model. The model’s performance was evaluated using AUC, precision, accuracy, recall, and F1-score and compared with other models. The rule-based diagnostic model exhibited excellent performance in the training, testing, and validation cohorts, with AUC values of 0.9621, 0.9529, and 0.8953, respectively. This model outperformed counterparts relying solely on selected features and previous research models. Specifically, the AUC values for the previous research models in the three cohorts were 0.851, 0.893, and 0.836. It is noteworthy that individual models employing GBDT, random forest, and AdaBoost demonstrated AUC values of 0.9391, 0.8681, and 0.9449 in the training cohort, 0.9093, 0.8722, and 0.9363 in the testing cohort, and 0.8440, 0.8640, and 0.8750 in the validation cohort, respectively. These results highlight the superiority of the rule-based diagnostic model in the assessment of lung adenocarcinoma subtypes, while also providing insights into the performance of individual models. Integrating diverse decision rules enhanced the accuracy and interpretability of the diagnostic model for lung adenocarcinoma subtypes. This approach bridges the gap between complex predictive models and clinical utility, offering valuable support to healthcare professionals and patients.

Generalizability of machine learning models predicting 30-day unplanned readmission after primary total knee arthroplasty using a nationally representative database.

Unplanned readmission after primary total knee arthroplasty (TKA) costs an average of US $39,000 per episode and negatively impacts patient outcomes. Although predictive machine learning (ML) models show promise for risk stratification in specific populations, existing studies do not address model generalizability. This study aimed to establish the generalizability of previous institutionally developed ML models to predict 30-day readmission following primary TKA using a national database. Data from 424,354 patients from the ACS-NSQIP database was used to develop and validate four ML models to predict 30-day readmission risk after primary TKA. Individual model performance was assessed and compared based on discrimination, accuracy, calibration, and clinical utility. Length of stay (> 2.5days), body mass index (BMI) (> 33.21kg/m2), and operation time (> 93min) were important determinants of 30-day readmission. All ML models demonstrated equally good accuracy, calibration, and discriminatory ability (Brier score, ANN = RF = HGB = NEPLR = 0.03; ANN, slope = 0.90, intercept = - 0.11; RF, slope = 0.93, intercept = - 0.12; HGB, slope = 0.90, intercept = - 0.12; NEPLR, slope = 0.77, intercept = 0.01; AUCANN = AUCRF = AUCHGB = AUCNEPLR = 0.78). This study validates the generalizability of four previously developed ML algorithms in predicting readmission risk in patients undergoing TKA and offers surgeons an opportunity to reduce readmissions by optimizing discharge planning, BMI, and surgical efficiency.

Comparison of transfer-learning for lightweight pre-trained model on image classification

This paper presents a comparative study of the performance of three convolutional neural network (CNN) architectures - EffcientNet-B0, ResNet-50, and AlexNet - for a given image classification task. The study provides a comprehensive investigation of the training process, hardware configurations, training time, and individual model performance. The investigation also assesses the models suitability for different applications. The findings can help both researchers and practitioners select the most suitable model for their specific needs and applications. The paper provides an analysis of each CNN architecture and discusses their strength and weaknesses. The results demonstrate that EffcientNet-B0 achieves the highest accuracy, but its training performance is not optimal. ResNet-50, on the other hand, exhibits high accuracy with efficient training using transfer learning. Finally, ALEXNET provides a baseline for comparison with traditional CNN designs. The paper also highlights the trade-offs involved in selecting a CNN architecture and highlights their relative advantages and disadvantages. The reader is provided with insights into which CNN architecture is most suitable for specific applications based on their requirements.

Prediction of customer demand for perishable products in retail inventory management, using the hybrid prophet-XGBoost model during the post-COVID-19 period

ABSTRACT Retail inventory management (IM) presents challenges in decision-making, especially in determining optimal inventory levels for future customer demand alignment. This study focuses on predicting dairy product demand by a significant European retail company using historical order data. As dairy products are perishable and constitute a relevant portion of retail trade volume, accurate IM is crucial to prevent wastage and meet customer needs. However, external factors like COVID-19 may impact demand volatility and seasonal patterns. Given the absence of scientific studies on predicting post-COVID-19 demand, our research aims to fill this gap by proposing alternatives, such as excluding the pandemic period and developing a historical imputation. The study employs two predictive models, Prophet and XGBoost, known for their superior performance in predicting demand. Moreover, a hybrid approach combining both models is proposed to enhance prediction accuracy by leveraging the capacity of the Prophet model to handle seasonal and holiday-period effects and XGBoost’s regularization to prevent overfitting. The results demonstrate the feasibility of historical imputation and the hybrid model approach, improving significantly individual model performance. The principal application of the study is to propose an approach to predict the shipment of other products in a post-COVID-19 context.

Highly Imbalanced Classification of Gout Using Data Resampling and Ensemble Method

Gout is one of the most painful diseases in the world. Accurate classification of gout is crucial for diagnosis and treatment which can potentially save lives. However, the current methods for classifying gout periods have demonstrated poor performance and have received little attention. This is due to a significant data imbalance problem that affects the learning attention for the majority and minority classes. To overcome this problem, a resampling method called ENaNSMOTE-Tomek link is proposed. It uses extended natural neighbors to generate samples that fall within the minority class and then applies the Tomek link technique to eliminate instances that contribute to noise. The model combines the ensemble ’bagging’ technique with the proposed resampling technique to improve the quality of generated samples. The performance of individual classifiers and hybrid models on an imbalanced gout dataset taken from the electronic medical records of a hospital is evaluated. The results of the classification demonstrate that the proposed strategy is more accurate than some imbalanced gout diagnosis techniques, with an accuracy of 80.87% and an AUC of 87.10%. This indicates that the proposed algorithm can alleviate the problems caused by imbalanced gout data and help experts better diagnose their patients.

Optimizing ensembles machine learning, genetic algorithms, and multivariate modeling for enhanced prediction of maize yield and stress tolerance index

The frequent occurrence of drought, halting from unpredictable climate-induced weather patterns, presents significant challenges in breeding drought-tolerant maize to identify adaptable genotypes. The study explores the optimization of machine learning (ML) to predict both the grain yield and stress tolerance index (STI) of maize under normal and drought-induced stress. In total, 35 genotypes, comprising 31 hybrid candidates and four commercial varieties, were meticulously evaluated across three normal and drought-treated sites. Three popular ML were optimized using a genetic algorithm (GA) and ensemble ML to enhance data capture. Additionally, a Multi-trait Genotype-Ideotype Distance (MGIDI) was also involved to identify superior maize hybrids well-suited for drought conditions. The results highlight that the ensemble meta-models optimized by grid search exhibit robust performance with high accuracy across the testing datasets (R2 = 0.92 for grain yield and 0.82 for STI). The RF optimized by GA algorithm demonstrates slightly lower performance (R2 = 0.91 for grain yield and 0.79 for STI), surpassing the predictive performance of individual SVM-GA and KNN-GA models. Selection of the best-performing hybrids indicated that out of the six hybrids with the highest STI values, both the ensemble and MGIDI can accurately predict four hybrids, namely H06, H10, H13, and H35. Thus, combining ML with MGIDI enables researchers to discern traits for each genotype and holds promise for advancing the field of drought-tolerant maize breeding and expediting the development of resilient varieties.

Biogas Production Prediction Based on Feature Selection and Ensemble Learning

The allocation of biogas between power generation and heat supply in traditional kitchen waste power generation system is unreasonable; for this reason, a biogas prediction method based on feature selection and heterogeneous model integration learning is proposed for biogas production predictions. Firstly, the working principle of the biogas generation system based on kitchen waste is analyzed, the relationship between system features and biogas production is mined, and the important features are extracted. Secondly, the prediction performance of different individual learner models is comprehensively analyzed, and the training set is divided to reduce the risk of overfitting by combining K-fold cross-validation. Finally, different primary learners and meta learners are selected according to the prediction error and diversity index, and different learners are fused to construct the stacking ensemble learning model with a two-layer structure. The experimental results show that the research method has a higher prediction accuracy in predicting biogas production, which provides supporting data for the economic planning of kitchen waste power generation systems.

Forecasting value-at-risk and expected shortfall in emerging market: does forecast combination help?

PurposeThis paper investigates how various strategies for combining forecasts, both simple and optimised approaches, are compared with popular individual risk models in estimating value-at-risk (VaR) and expected shortfall (ES) in emerging market at alternative risk levels.Design/methodology/approachUsing the case study of the Vietnamese stock market, the author produced one-day-ahead VaR and ES forecast from seven individual risk models and ten alternative forecast combinations. Next, the author employed a battery of backtesting procedures and alternative loss functions to evaluate the global predictive accuracy of the different methods. Finally, the author investigated the relative performance over time of VaR and ES forecasts using fluctuation test.FindingsThe empirical results indicate that, although combined forecasts have reasonable predictive abilities, they are often outperformed by one individual risk model. Furthermore, the author showed that the complex combining methods with optimised weighting functions do not perform better than simple combining methods. The fluctuation test suggests that the poor performance of combined forecasts is mainly due to their inability to cope with periods of instability.Research limitations/implicationsThis study reveals the limitation of combining strategies in the one-day-ahead VaR and ES forecasts in emerging markets. A possible direction for further research is to investigate whether this finding holds for multi-day ahead forecasts. Moreover, the inferior performance of combined forecasts during periods of instability motivates further research on the combining strategies that take into account for potential structure breaks in the performance of individual risk models. A potential approach is to improve the individual risk models with macroeconomic variables using a mixed-data sampling approach.Originality/valueFirst, the authors contribute to the literature on the forecasting combinations for VaR and ES measures. Second, the author explored a wide range of alternative risk models to forecast both VaR and ES with recent data including periods of the COVID-19 pandemic. Although forecast combination strategies have been providing several good results in several fields, the literature of forecast combination in the VaR and ES context is surprisingly limited, especially for emerging market returns. To the best of the author’s knowledge, this is the first study investigating predictive power of combining methods for VaR and ES in an emerging market.

Forest Fire Risk Prediction Based on Stacking Ensemble Learning for Yunnan Province of China

Forest fire risk prediction is essential for building a forest fire defense system. Ensemble learning methods can avoid the problem of difficult model selection for disaster susceptibility prediction and can significantly improve modeling accuracy. This study introduces a stacking ensemble learning model for predicting forest fire risks in Yunnan Province by integrating various data types, such as meteorological, topographic, vegetation, and human activity factors. A total of 70,274 fire points and an equal number of randomly selected nonfire points were used to develop the model, with 70% of the data allocated for training and the remaining 30% for testing. The stacking model combined four diverse machine learning methods: random forest (RF), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and multilayer perceptron (MLP). We evaluated the model’s predictive performance using metrics like accuracy, area under the characteristic curve (AUC), and fire density (FD). The results demonstrated that the stacking fusion model exhibited remarkable accuracy with an AUC of 0.970 on the test set, significantly surpassing the performance of individual machine learning models, which had AUC values ranging from 0.935 to 0.953. Furthermore, the stacking fusion model effectively captured the maximum fire density in extremely high susceptibility areas, demonstrating enhanced generalization capabilities.