Light Gradient Boosting Machine Research Articles

Predictive model and risk analysis for coronary heart disease in people living with HIV using machine learning

ObjectiveThis study aimed to construct a coronary heart disease (CHD) risk-prediction model in people living with human immunodeficiency virus (PLHIV) with the help of machine learning (ML) per electronic medical records (EMRs).MethodsSixty-one medical characteristics (including demography information, laboratory measurements, and complicating disease) readily available from EMRs were retained for clinical analysis. These characteristics further aided the development of prediction models by using seven ML algorithms [light gradient-boosting machine (LightGBM), support vector machine (SVM), eXtreme gradient boosting (XGBoost), adaptive boosting (AdaBoost), decision tree, multilayer perceptron (MLP), and logistic regression]. The performance of this model was assessed using the area under the receiver operating characteristic curve (AUC). Shapley additive explanation (SHAP) was further applied to interpret the findings of the best-performing model.ResultsThe LightGBM model exhibited the highest AUC (0.849; 95% CI, 0.814–0.883). Additionally, the SHAP plot per the LightGBM depicted that age, heart failure, hypertension, glucose, serum creatinine, indirect bilirubin, serum uric acid, and amylase can help identify PLHIV who were at a high or low risk of developing CHD.ConclusionThis study developed a CHD risk prediction model for PLHIV utilizing ML techniques and EMR data. The LightGBM model exhibited improved comprehensive performance and thus had higher reliability in assessing the risk predictors of CHD. Hence, it can potentially facilitate the development of clinical management techniques for PLHIV care in the era of EMRs.

Use Machine Learning to Forecast Economic Recession with Covid-19 Evidence

This article starts by studying the impact of the COVID-19 pandemic on the United States economy and then delves into the relationship between the pandemic and the economic mobility in U.S. In addition, this article also uses various models to predict the US economy, including vector autoregressive model, support vector machines, eXtreme gradient boosting, light gradient-boosting machine and the long short-term memory network, then comparing the prediction results of the above models to select the relatively optimal model. To test the model, the model is also used to predict Italy's economy and then XGBoost is selected. The limitations of the model will be proposed based on the predicted results. Based on the research, XGBoost model can be applied to forecast economic recession. With further study, we believe such forecast and our understanding of economic processes and economic recession can be improved, and the government can adopt proper policies to alleviate the economic recession.

A maneuver indicator and ensemble learning-based risky driver recognition approach for highway merging areas

Abstract Due to the complex traffic characteristics in highway merging areas, drivers tend to exhibit high-risk driving behaviors. To address the characteristics of driving behavior in highway merging areas, we have developed a real-time identification model for risky drivers by combining a driver risk level labeling method with load balancing-ensemble learning (LB-EL). In this paper, we explore four types of maneuver indicator indexes (MIIs)—acute direction, stomp pedal, dangerous following, and dangerous lane changing—that can describe the negative behaviors of both individual vehicles and vehicle platoons in highway merging areas. To quantize the label driver risk level, we use the interquartile range (IQR) method and Criteria Importance Though Intercriteria Correlation (CRITIC), while we evaluate the reliability of the MII using spatial analysis. Furthermore, we balance the dataset using three load balancing (LB) algorithms and create nine ensemble strategies by pairing adaptive boosting (AdaBoost), extreme gradient boosting (XGBoost), and light gradient boosting machine (LGBM) with the three LB algorithms. Finally, we validate the proposed model using trajectory data extracted from UAV videos. The results indicate that the distribution laws of risky driving behaviors in the acute direction and stomp pedal show a high degree of similarity and good matching with the distribution laws of traffic conflict points in existing research. Moreover, the SMOTE-LGBM ensemble model achieves the best performance, reaching an accuracy rate of 93.4% and a recall rate of 92.1%, which demonstrates the validity of our proposed model. This model can be widely applied to recognize risky drivers in video-based surveillance systems.

Advanced forecasting of COVID-19 epidemic: Leveraging ensemble models, advanced optimization, and decomposition techniques

In the global effort to address the outbreak of the new coronavirus pneumonia (COVID-19) pandemic, accurate forecasting of epidemic patterns has become crucial for implementing successful interventions aimed at preventing and controlling the spread of the disease. The correct prediction of the course of COVID-19 outbreaks is a complex and challenging task, mainly because of the significant volatility in the data series related to COVID-19. Previous studies have been limited by the exclusive use of individual forecasting techniques in epidemic modeling, disregarding the integration of diverse prediction procedures. The lack of attention to detail in this situation can yield worse-than-ideal results. Consequently, this study introduces a novel ensemble framework that integrates three machine learning methods (kernel ridge regression (KRidge), Deep random vector functional link (dRVFL), and ridge regression) within a linear relationship (L-KRidge-dRVFL-Ridge). The optimization of this framework is accomplished through a distinctive approach, specifically adaptive differential evolution and particle swarm optimization (A-DEPSO). Moreover, an effective decomposition method, known as time-varying filter empirical mode decomposition (TVF-EMD), is employed to decompose the input variables. A feature selection technique, specifically using the light gradient boosting machine (LGBM), is also implemented to extract the most influential input variables. The daily datasets of COVID-19 collected from two countries, namely Italy and Poland, were used as the experimental examples. Additionally, all models are implemented to forecast COVID-19 at two-time horizons: 10- and 14-day ahead (t+10 and t+14). According to the results, the proposed model can yield higher correlation coefficient (R) for both case studies: Italy (t+10 = 0.965, t+14 = 0.961) and Poland (t+10 = 0.952, t+14 = 0.940) than the other models. The experimental results demonstrate that the model suggested in this paper has outstanding results in various kinds of complex epidemic prediction situations. The proposed ensemble model demonstrates exceptional accuracy and resilience, outperforming all similar models in terms of efficacy.

Identifying the top predictors of student well-being across cultures using machine learning and conventional statistics

Alongside academic learning, there is increasing recognition that educational systems must also cater to students’ well-being. This study examines the key factors that predict adolescent students’ subjective well-being, indexed by life satisfaction, positive affect, and negative affect. Data from 522,836 secondary school students from 71 countries/regions across eight different cultural contexts were analyzed. Underpinned by Bronfenbrenner’s bioecological theory, both machine learning (i.e., light gradient-boosting machine) and conventional statistics (i.e., hierarchical linear modeling) were used to examine the roles of person, process, and context factors. Among the multiple predictors examined, school belonging and sense of meaning emerged as the common predictors of the various well-being dimensions. Different well-being dimensions also had distinct predictors. Life satisfaction was best predicted by a sense of meaning, school belonging, parental support, fear of failure, and GDP per capita. Positive affect was most strongly predicted by resilience, sense of meaning, school belonging, parental support, and GDP per capita. Negative affect was most strongly predicted by fear of failure, gender, being bullied, school belonging, and sense of meaning. There was a remarkable level of cross-cultural similarity in terms of the top predictors of well-being across the globe. Theoretical and practical implications are discussed.

Fast Healthcare Interoperability Resources-Based Support System for Predicting Delivery Type: Model Development and Evaluation Study.

The escalating prevalence of cesarean delivery globally poses significant health impacts on mothers and newborns. Despite this trend, the underlying reasons for increased cesarean delivery rates, which have risen to 36.3% in Portugal as of 2020, remain unclear. This study delves into these issues within the Portuguese health care context, where national efforts are underway to reduce cesarean delivery occurrences. This paper aims to introduce a machine learning, algorithm-based support system designed to assist clinical teams in identifying potentially unnecessary cesarean deliveries. Key objectives include developing clinical decision support systems for cesarean deliveries using interoperability standards, identifying predictive factors influencing delivery type, assessing the economic impact of implementing this tool, and comparing system outputs with clinicians' decisions. This study used retrospective data collected from 9 public Portuguese hospitals, encompassing maternal and fetal data and delivery methods from 2019 to 2020. We used various machine learning algorithms for model development, with light gradient-boosting machine (LightGBM) selected for deployment due to its efficiency. The model's performance was compared with clinician assessments through questionnaires. Additionally, an economic simulation was conducted to evaluate the financial impact on Portuguese public hospitals. The deployed model, based on LightGBM, achieved an area under the receiver operating characteristic curve of 88%. In the trial deployment phase at a single hospital, 3.8% (123/3231) of cases triggered alarms for potentially unnecessary cesarean deliveries. Financial simulation results indicated potential benefits for 30% (15/48) of Portuguese public hospitals with the implementation of our tool. However, this study acknowledges biases in the model, such as combining different vaginal delivery types and focusing on potentially unwarranted cesarean deliveries. This study presents a promising system capable of identifying potentially incorrect cesarean delivery decisions, with potentially positive implications for medical practice and health care economics. However, it also highlights the challenges and considerations necessary for real-world application, including further evaluation of clinical decision-making impacts and understanding the diverse reasons behind delivery type choices. This study underscores the need for careful implementation and further robust analysis to realize the full potential and real-world applicability of such clinical support systems.

Development and validation of machine learning models to predict postoperative infarction in moyamoya disease.

Cerebral infarction is a common complication in patients undergoing revascularization surgery for moyamoya disease (MMD). Although previous statistical evaluations have identified several risk factors for postoperative brain ischemia, the ability to predict its occurrence based on these limited predictors remains inadequately explored. This study aimed to assess the feasibility of machine learning algorithms for predicting cerebral infarction after revascularization surgery in patients with MMD. This retrospective study was conducted across two centers and harnessed data from 512 patients with MMD who had undergone revascularization surgery. The patient cohort was partitioned into internal and external datasets. Using perioperative clinical data from the internal cohort, three distinct machine learning algorithms-namely the support vector machine, random forest, and light gradient-boosting machine models-were trained and cross-validated to predict the occurrence of postoperative cerebral infarction. Predictive performance validity was subsequently assessed using an external dataset. Shapley additive explanations (SHAP) analysis was conducted to augment the prediction model's transparency and to quantify the impact of each input variable on shaping both the aggregate and individual patient predictions. In the cohort of 512 patients, 33 (6.4%) experienced postrevascularization cerebral infarction. The cross-validation outcomes revealed that, among the three models, the support vector machine model achieved the largest area under the receiver operating characteristic curve (ROC-AUC) at mean ± SD 0.785 ± 0.052. Notably, during external validation, the light gradient-boosting machine model exhibited the highest accuracy at 0.903 and the largest ROC-AUC at 0.710. The top-performing prediction model utilized five input variables: postoperative serum gamma-glutamyl transpeptidase value, positive posterior cerebral artery (PCA) involvement on preoperative MRA, infarction as the rationale for surgery, presence of an infarction scar on preoperative MRI, and preoperative modified Rankin Scale score. Furthermore, the SHAP analysis identified presence of PCA involvement, infarction as the rationale for surgery, and presence of an infarction scar on preoperative MRI as positive influences on postoperative cerebral infarction. This study indicates the usefulness of employing machine learning techniques with routine perioperative data to predict the occurrence of cerebral infarction after revascularization procedures in patients with MMD.

Musical instrument classifier for early childhood percussion instruments.

While the musical instrument classification task is well-studied, there remains a gap in identifying non-pitched percussion instruments which have greater overlaps in frequency bands and variation in sound quality and play style than pitched instruments. In this paper, we present a musical instrument classifier for detecting tambourines, maracas and castanets, instruments that are often used in early childhood music education. We generated a dataset with diverse instruments (e.g., brand, materials, construction) played in different locations with varying background noise and play styles. We conducted sensitivity analyses to optimize feature selection, windowing time, and model selection. We deployed and evaluated our best model in a mixed reality music application with 12 families in a home setting. Our dataset was comprised of over 369,000 samples recorded in-lab and 35,361 samples recorded with families in a home setting. We observed the Light Gradient Boosting Machine (LGBM) model to perform best using an approximate 93 ms window with only 12 mel-frequency cepstral coefficients (MFCCs) and signal entropy. Our best LGBM model was observed to perform with over 84% accuracy across all three instrument families in-lab and over 73% accuracy when deployed to the home. To our knowledge, the dataset compiled of 369,000 samples of non-pitched instruments is first of its kind. This work also suggests that a low feature space is sufficient for the recognition of non-pitched instruments. Lastly, real-world deployment and testing of the algorithms created with participants of diverse physical and cognitive abilities was also an important contribution towards more inclusive design practices. This paper lays the technological groundwork for a mixed reality music application that can detect children's use of non-pitched, percussion instruments to support early childhood music education and play.

A Real‐time Lithological Identification Method based on SMOTE‐Tomek and ICSA Optimization

AbstractIn petroleum engineering, real‐time lithology identification is very important for reservoir evaluation, drilling decisions and petroleum geological exploration. A lithology identification method while drilling based on machine learning and mud logging data is studied in this paper. This method can effectively utilize downhole parameters collected in realtime during drilling, to identify lithology in real‐time and provide a reference for optimization of drilling parameters. Given the imbalance of lithology samples, the synthetic minority over‐sampling technique (SMOTE) and Tomek link were used to balance the sample number of five lithologies. Meanwhile, this paper introduces Tent map, random opposition‐based learning and dynamic perceived probability to the original crow search algorithm (CSA), and establishes an improved crow search algorithm (ICSA). In this paper, ICSA is used to optimize the hyperparameter combination of random forest (RF), extremely random trees (ET), extreme gradient boosting (XGB), and light gradient boosting machine (LGBM) models. In addition, this study combines the recognition advantages of the four models. The accuracy of lithology identification by the weighted average probability model reaches 0.877. The study of this paper realizes high‐precision real‐time lithology identification method, which can provide lithology reference for the drilling process.

The drug loading capacity prediction and cytotoxicity analysis of metal–organic frameworks using stacking algorithms of machine learning

Metal–organic frameworks (MOFs) have shown excellent performance in the field of drug delivery. Despite the synthesis of a vast array of MOFs exceeding 100,000 varieties, certain formulations have exhibited suboptimal performance characteristics. Therefore, there is a pressing need to enhance their efficacy by identifying MOFs with superior drug loading capacities and minimal cytotoxicity, which can be achieved through machine learning (ML). In this study, a stacking regression model was developed to predict drug loading capacity and cytotoxicity of MOFs using datasets compiled from various literature sources. The model exhibited exceptional predictive capabilities, achieving R2 values of 0.907 for drug loading capacity and 0.856 for cytotoxicity. Furthermore, various model interpretation methods including partial dependence plots, individual conditional expectation, Shapley additive explanation, decision tree, random forest, CatBoost Regressor, and light gradient-boosting machine were employed for feature importance analysis. The results revealed that specific metal atoms such as Zn, Cr, Fe, Zr, and Cu significantly influenced the drug loading capacity and cytotoxicity of MOFs. Through model validation encompassing experimental validation and computational verification, the reliability of the model was thoroughly established. In general, it is a good practice to use ML methods for predicting drug loading capacity and cytotoxicity analysis of MOFs, guiding the development of future property prediction methods for MOFs.

Landslide Susceptibility Zoning: Integrating Multiple Intelligent Models with SHAP Analysis

In this study, we aim to delineate landslide susceptibility zones within Dien Bien province, Vietnam, leveraging the capabilities of various machine learning models including Light Gradient Boosting Machine (LGBM), K-Nearest Neighbors (KNN), and Gradient Boosting (GB). Harnessing a dataset comprising 665 data points and encompassing 14 influential factors such as slope, aspect, curvature, elevation, geological composition, Normalized Difference Vegetation Index (NDVI), and proximity to geological features like faults, rivers, and roads, a comprehensive database for landslide modeling was constructed. The analysis entailed rigorous evaluation and comparison of model accuracy employing established statistical metrics, notably Receiver Operating Characteristic (ROC) curves and Area Under the Curve (AUC). The findings underscore the efficacy of the Light Gradient Boosting Machine model, exhibiting superior performance with an AUC score of 0.85, surpassing both the Gradient Boosting model (AUC = 0.81) and the K-Nearest Neighbors model (AUC = 0.79). Notably, the Light Gradient Boosting Machine model emerges as a promising tool for precise landslide prediction within the study area, offering significant potential for the creation of a robust landslide susceptibility map. The resulting spatial forecast map for Dien Bien province holds considerable utility for informing land use planning initiatives aimed at mitigating the impact of landslide disasters in the region. Moreover, the application of SHAP (Shapley Additive explanation) values to quantify the contribution of each factor to landslide susceptibility prediction, offering novel insights into model interpretation and feature importance. The resulting spatial forecast map holds significant implications for land use planning and disaster mitigation efforts in Dien Bien province, showcasing the potential of advanced machine learning techniques in enhancing landslide risk management strategies.

Explainable prediction of problematic smartphone use among South Korea's children and adolescents using a Machine learning approach

BackgroundKorea is known for its technological prowess, has the highest smartphone ownership rate in the world at 95%, and the smallest gap in smartphone ownership between generations. Since the onset of the COVID-19 pandemic, problematic smartphone use is becoming more prevalent among Korean children and adolescent owing to limited school attendance and outdoor activities, resulting in increased reliance on smartphones. 40.1% of adolescents are classified as high-risk, with only the adolescent group showing a persistent rise year after year. ObjectiveThe study purpose is to present data-driven analysis results for predicting and preventing smartphone addiction in Korea, where problematic smartphone use is severe.Participants and Methods: To predict the risk of problematic smartphone use in Korean children and adolescents at an early stage, we used data collected from the Smartphone Overdependence Survey conducted by the National Information Society Agency between 2017 and 2021. Eight representative machine and deep learning algorithms were used to predict groups at high risk for smartphone addiction: Logistic Regression, Random Forest, Gradient Boosting Machine (GBM), extreme Gradient Boosting (XGBoost), Light GBM, Categorical Boosting, Multilayer Perceptron, and Convolutional Neural Network. ResultsThe XGBoost ensemble algorithm predicted 87.60% of participants at risk of future problematic smartphone usebased on precision. Our results showed that prolonged use of games, webtoons/web novels, and e-books, which have not been found in previous studies, further increased the risk of problematic smartphone use. ConclusionsArtificial intelligence algorithms have potential predictive and explanatory capabilities for identifying early signs of problematic smartphone use in adolescents and young children. We recommend that a variety of healthy, beneficial, and face-to-face activities be offered as alternatives to smartphones for leisure and play culture.

A novel deep Siamese framework for burned area mapping Leveraging mixture of experts

Due to the complexity of the areas and the diversity of the objects, traditional Burned Area Mapping (BAM) methods cannot provide promising results. Moreover, these methods focus on additional processing to improve their results, which is time-consuming and complex. Therefore, an advanced framework is needed to achieve accurate results in burned area mapping. In this context, this study proposes a novel Siamese-based mixture of expert networks for burned area mapping, which takes advantage of a mixture of experts (MoE) and position and channel attention mechanisms for deep feature generation. The proposed framework has two deep feature extractor channels to account for the bi-temporal multispectral pre- and post-fire input datasets. To evaluate the performance of the SMoE model, three multispectral Sentinel-2 were used in different countries. The results were compared with other advanced machine and deep learning models, including Light Gradient Boosting Machine (LGBM), Extreme Gradient Boosting (XGBoost), 2D-Siamse Convolutional Neural Network (2D-SCNN), and 3D-Siamse Convolutional Neural Network (3D-SCNN). The result of the burned mapping shows that the proposed model has a high effectiveness compared to other models, as it provides an average accuracy of more than 98% and 0.91 b y overall accuracy (OA) and kappa coefficient (KC) indices, respectively.

Identification of the best machine learning model for the prediction of driver injury severity

Predicting the injury severities sustained by drivers engaged in road traffic accidents is a key topic of research in road traffic safety. The current study analyzed the driver injury severity (DIS) using twelve machine learning (ML) algorithms. These models were implemented using 0.70, 0.80, and 0.90 train ratios and 5-, 10- and 15-fold cross-validation. Ten years of accident data (from 2011 to 2020) was obtained from police department of Shillong, India. A total of 693 accidents were documented, with 68% being nonfatal and 32% being fatal. Precision, recall, accuracy, F1 score and area under the curve measures were used to compare the performance of all twelve ML models. Overall, the light gradient-boosting machine model was shown to be the best ML model for predicting the injury severities of drivers engaged in road traffic incidents. Finally, variable importance analysis results showed that cause of accident, collision type and types of vehicles were the most influencing factors in nonfatal and fatal driver accidents. The results also revealed that age and gender were slightly associated with DIS. The findings of the current research could be helpful to road safety agencies for the implementation of suitable countermeasures to increase driver safety in road accidents.

A predictive model for post-thoracoscopic surgery pulmonary complications based on the PBNN algorithm

We constructed an early prediction model for postoperative pulmonary complications after thoracoscopic surgery using machine learning and deep learning algorithms. The artificial intelligence prediction models were built in Python, primarily using artificial intelligencealgorithms including both machine learning and deep learning algorithms. Correlation analysis showed that postoperative pulmonary complications were positively correlated with age and surgery duration, and negatively correlated with serum albumin. Using the light gradient boosting machine(LGBM) algorithm, weighted feature engineering revealed that single lung ventilation duration, history of smoking, surgery duration, ASA score, and blood glucose were the main factors associated with postoperative pulmonary complications. Results of artificial intelligence algorithms for predicting pulmonary complications after thoracoscopy in the test group: In terms of accuracy, the two best algorithms were Logistic Regression (0.831) and light gradient boosting machine(0.827); in terms of precision, the two best algorithms were Gradient Boosting (0.75) and light gradient boosting machine (0.742); in terms of recall, the three best algorithms were gaussian naive bayes (0.581), Logistic Regression (0.532), and pruning Bayesian neural network (0.516); in terms of F1 score, the two best algorithms were LogisticRegression (0.589) and pruning Bayesian neural network (0.566); and in terms of Area Under Curve(AUC), the two best algorithms were light gradient boosting machine(0.873) and pruning Bayesian neural network (0.869). The results of this study suggest that pruning Bayesian neural network (PBNN) can be used to assess the possibility of pulmonary complications after thoracoscopy, and to identify high-risk groups prior to surgery.

Comparison of machine learning algorithms and feature extraction techniques for the automatic detection of surface EMG activation timing

This paper presents a methodology for automatically detecting muscular activity by denoising, extracting features, and classifying surface electromyography (sEMG) signals. The proposed methodology utilizes the Discrete Wavelet Transform (DWT) and Willison’s Amplitude Algorithm (WAMP) for feature extraction. Five classification methods, including Neural Networks (NN), Classification Vector, XGBoost, Light Gradient Boosting Machine (LGBM), and ExtraTree, were evaluated using F-Measure, Precision, and Recall as performance metrics. Through k-fold cross-validation, the XGBoost algorithm, when combined with the Eigen values feature, achieved the highest training performance with an F1-Score of 98.71 %. For the test group, the LGBM classifier using WAMP, and NN with both WAMP and Eigen values as features, demonstrated the best average performance with F1-Scores of 96.52 ± 3.45 % and 96.52 ± 3.07 %, respectively. These results highlight the precision and performance of the proposed approach in detecting EMG signals. Moreover, the framework has the potential to support clinicians in diagnosing neuromuscular disorders and developing human–machine interfaces.

Unleashing the power of machine learning and remote sensing for robust seasonal drought monitoring: A stacking ensemble approach

Droughts cause significant economic losses in many regions around the world, highlighting a need to more accurately quantify implications of drought on production and water management. Remote sensing technologies and machine learning-based models offer promising solutions for timely and accurate regional drought monitoring, but the accuracy of such approaches is constrained by both the conceptual design and algorithms underpinning such approaches. We developed a machine learning stacking ensemble approach to overcome such limitations, drawing upon Precipitation Estimation from Remotely Sensed Information using Artificial Neural Network-Climate Data Record (PERSIANN-CDR), Moderate Resolution Imaging Spectroradiometer (MODIS) remote sensing products, and climate zoning data to estimate 3-month scale SPEI (Standardized Precipitation Evapotranspiration Index, SPEI-3) in 9 sub-regions of China. We compared 19 individual machine learning models and used the stacking approach to select the most robust meta-model for SPEI-3 prediction. We found that CatBoost Regressor (CBR), Extra Trees Regressor (ETR), Extreme Gradient Boosting (XGB), Light Gradient Boosting Machine (LGBM), and Random Forest (RF) were the top individual models for predicting drought, while using CBR as the stacked meta-model achieved the best performance. The R2 values for the stacking model with CBR as the meta-model were 0.9065 and 0.8218 in the eastern and western regions, respectively. We then employed the stacking model with CBR as the meta-model to generate seasonal drought maps based on SPEI-3 across different years and seasons. Our predicted SPEI-3 was compared with drought maps generated by the GPCC and ERA5 reanalysis datasets. We found a strong correlation between them with R2 exceeding 0.8 across multiple years and seasons, indicating our machine learning stacking ensemble approach had a good performance in monitoring seasonal drought conditions. The findings of our study establish a standardized protocol for predicting droughts in the agricultural sector, enabling its widespread application irrespective of variations in terrain and climate.

Application of machine learning in prediction of Pb2+ adsorption of biochar prepared by tube furnace and fluidized bed.

Data mining by machine learning (ML) has recently come into application in heavy metals purification from wastewater, especially in exploring lead removal by biochar that prepared using tube furnace (TF-C) and fluidized bed (FB-C) pyrolysis methods. In this study, six ML models including Random Forest Regression (RFR), Gradient Boosting Regression (GBR), Support Vector Regression (SVR), Kernel Ridge Regression (KRR), Extreme Gradient Boosting (XGB), and Light Gradient Boosting Machine (LGBM) were employed to predict lead adsorption based on a dataset of 1012 adsorption experiments, comprising 422 TF-C groups from our experiments and 590 FB-C groups from literatures. The XGB model showed superior accuracy and predictive performance for adsorption, achieving R2 values for TF-C (0.992) and FB-C (0.981), respectively. Contrasting inferior results were observed in other models, including RF (0.962 and 0.961), GBR (0.987 and 0.975), SVR (0.839 and 0.763), KRR (0.817 and 0.881), and LGBM (0.975 and 0.868). Additionally, a hybrid dataset combining both biochars in Pb adsorption also indicated high accuracy (0.972) as obtained from XGB model. The investigation revealed that the influence of char characteristics and adsorption conditions on Pb adsorption differs between the two biochar. Specific char characteristics, particularly nitrogen content, significantly influence lead adsorption in both biochar. Interestingly, the influence of pyrolysis temperature (PT) on lead adsorption is found to be greater for TF-C than for FB-C. Consequently, careful consideration of PT is crucial when preparing TF-C biochar. These findings offer practical guidance for optimizing biochar preparation conditions during heavy metal removal from wastewater.

PM2.5 concentration forecasting: Development of integrated multivariate variational mode decomposition with kernel Ridge regression and weighted mean of vectors optimization

The accurate prediction of the PM2.5 air quality parameter within industrial and urban settings is the most pressing issue examined by the researchers because it has high health implications. However, accurately forecasting PM2.5 levels is crucial. The traditional machine learning (ML) models are incapable, as the indices fluctuate daily. To successfully manage this problem, a new ML framework is proposed that incorporates various techniques, such as LGBM feature selection (light gradient-boosting machine), MVMD (multivariate variational mode decomposition), KRidge (kernel Ridge regression), and INFO (weighted mean of vectors). The proposed framework is used to estimate PM2.5 pollution at specific stations in China for a one-time prediction. The LGBM feature selection technique is the first step of the pre-processing, which performs to select the most important variables. Next, the MVMD splits the initial signal into intrinsic mode functions (IMFs), accommodating the signal's non-stationary multivariate. Following that, the KRidge approach is applied to each sub-component using the best input feature, and the resulting predictions are summed up to get the PM2.5 levels. To assess the validity of the proposed MVMD-KRidge-INFO model, the categories of gaussian-process-regression (GPR), locally-weighted-linear-regression (LWLR), and multivariate adaptive regression splines (MARS) are analyzed in individual and hybrid moldes. As observed from the research results, MVMD-KRidge-INFO performs optimally at forecasting PM2.5 levels at Huairou and Shunyi, as evidenced by R, RMSE, MAPE, IA, MdAE, and U95% metrics.

Effect of freeze–thaw cycles on membrane electrode assembly of proton exchange membrane fuel cells and its fault diagnosis method

AbstractIn low‐temperature environment, the residual water in the membrane electrode assembly (MEA) will freeze after the operation of proton exchange membrane fuel cells, which will cause damage to the MEA. In this paper, the effect of freeze–thaw cycles on MEA was studied. Six sets of MEA samples with 0, 20, 40, 60, 80, and 100 times freeze–thaw cycles were set up, and the damage on MEAs is analyzed by polarization curves, electrochemical impedance spectra, cyclic voltammetry curves, and scanning electron microscope. It was found that the freeze–thaw cycles caused degradation on MEA, and the ohmic resistance of MEA increases with the number of cycles increases before the 60 freeze–thaw cycles, and after 60 freeze–thaw cycles, a gap appeared between the proton exchange membrane (PEM) and the catalyst layer, which led to more water entering the PEM and the ohmic resistance of MEA decreased. Besides, according to the data analysis, the experimental samples are divided into three categories (normal MEA, lightly damaged MEA, and seriously damaged MEA). A classifier model combining inception network and light gradient boosting machine (LGBM) was established, and it was found that the combined model was better than inception–dense and LGBM for classification, reaching 96.89%.