To enhance the accuracy and efficiency of bridge damage identification,a novel data-driven damage identification method was proposed.First,convolutional autoencoder(CAE)was used to extract key features from the accele...To enhance the accuracy and efficiency of bridge damage identification,a novel data-driven damage identification method was proposed.First,convolutional autoencoder(CAE)was used to extract key features from the acceleration signal of the bridge structure through data reconstruction.The extreme gradient boosting tree(XGBoost)was then used to perform analysis on the feature data to achieve damage detection with high accuracy and high performance.The proposed method was applied in a numerical simulation study on a three-span continuous girder and further validated experimentally on a scaled model of a cable-stayed bridge.The numerical simulation results show that the identification errors remain within 2.9%for six single-damage cases and within 3.1%for four double-damage cases.The experimental validation results demonstrate that when the tension in a single cable of the cable-stayed bridge decreases by 20%,the method accurately identifies damage at different cable locations using only sensors installed on the main girder,achieving identification accuracies above 95.8%in all cases.The proposed method shows high identification accuracy and generalization ability across various damage scenarios.展开更多
BACKGROUND Development of distant metastasis(DM)is a major concern during treatment of nasopharyngeal carcinoma(NPC).However,studies have demonstrated im-proved distant control and survival in patients with advanced N...BACKGROUND Development of distant metastasis(DM)is a major concern during treatment of nasopharyngeal carcinoma(NPC).However,studies have demonstrated im-proved distant control and survival in patients with advanced NPC with the addition of chemotherapy to concomitant chemoradiotherapy.Therefore,precise prediction of metastasis in patients with NPC is crucial.AIM To develop a predictive model for metastasis in NPC using detailed magnetic resonance imaging(MRI)reports.METHODS This retrospective study included 792 patients with non-distant metastatic NPC.A total of 469 imaging variables were obtained from detailed MRI reports.Data were stratified and randomly split into training(50%)and testing sets.Gradient boosting tree(GBT)models were built and used to select variables for predicting DM.A full model comprising all variables and a reduced model with the top-five variables were built.Model performance was assessed by area under the curve(AUC).RESULTS Among the 792 patients,94 developed DM during follow-up.The number of metastatic cervical nodes(30.9%),tumor invasion in the posterior half of the nasal cavity(9.7%),two sides of the pharyngeal recess(6.2%),tubal torus(3.3%),and single side of the parapharyngeal space(2.7%)were the top-five contributors for predicting DM,based on their relative importance in GBT models.The testing AUC of the full model was 0.75(95%confidence interval[CI]:0.69-0.82).The testing AUC of the reduced model was 0.75(95%CI:0.68-0.82).For the whole dataset,the full(AUC=0.76,95%CI:0.72-0.82)and reduced models(AUC=0.76,95%CI:0.71-0.81)outperformed the tumor node-staging system(AUC=0.67,95%CI:0.61-0.73).CONCLUSION The GBT model outperformed the tumor node-staging system in predicting metastasis in NPC.The number of metastatic cervical nodes was identified as the principal contributing variable.展开更多
This paper aims to design an optimizer followed by a Kawahara filter for optimal classification and prediction of employees’performance.The algorithm starts by processing data by a modified K-means technique as a hie...This paper aims to design an optimizer followed by a Kawahara filter for optimal classification and prediction of employees’performance.The algorithm starts by processing data by a modified K-means technique as a hierarchical clustering method to quickly obtain the best features of employees to reach their best performance.The work of this paper consists of two parts.The first part is based on collecting data of employees to calculate and illustrate the performance of each employee.The second part is based on the classification and prediction techniques of the employee performance.This model is designed to help companies in their decisions about the employees’performance.The classification and prediction algorithms use the Gradient Boosting Tree classifier to classify and predict the features.Results of the paper give the percentage of employees which are expected to leave the company after predicting their performance for the coming years.Results also show that the Grasshopper Optimization,followed by“KF”with the Gradient Boosting Tree as classifier and predictor,is characterized by a high accuracy.The proposed algorithm is compared with other known techniques where our results are fund to be superior.展开更多
Traditional 3Ni weathering steel cannot completely meet the requirements for offshore engineering development,resulting in the design of novel 3Ni steel with the addition of microalloy elements such as Mn or Nb for st...Traditional 3Ni weathering steel cannot completely meet the requirements for offshore engineering development,resulting in the design of novel 3Ni steel with the addition of microalloy elements such as Mn or Nb for strength enhancement becoming a trend.The stress-assisted corrosion behavior of a novel designed high-strength 3Ni steel was investigated in the current study using the corrosion big data method.The information on the corrosion process was recorded using the galvanic corrosion current monitoring method.The gradi-ent boosting decision tree(GBDT)machine learning method was used to mine the corrosion mechanism,and the importance of the struc-ture factor was investigated.Field exposure tests were conducted to verify the calculated results using the GBDT method.Results indic-ated that the GBDT method can be effectively used to study the influence of structural factors on the corrosion process of 3Ni steel.Dif-ferent mechanisms for the addition of Mn and Cu to the stress-assisted corrosion of 3Ni steel suggested that Mn and Cu have no obvious effect on the corrosion rate of non-stressed 3Ni steel during the early stage of corrosion.When the corrosion reached a stable state,the in-crease in Mn element content increased the corrosion rate of 3Ni steel,while Cu reduced this rate.In the presence of stress,the increase in Mn element content and Cu addition can inhibit the corrosion process.The corrosion law of outdoor-exposed 3Ni steel is consistent with the law based on corrosion big data technology,verifying the reliability of the big data evaluation method and data prediction model selection.展开更多
The Darjeeling Himalayan region,characterized by its complex topography and vulnerability to multiple environmental hazards,faces significant challenges including landslides,earthquakes,flash floods,and soil loss that...The Darjeeling Himalayan region,characterized by its complex topography and vulnerability to multiple environmental hazards,faces significant challenges including landslides,earthquakes,flash floods,and soil loss that critically threaten ecosystem stability.Among these challenges,soil erosion emerges as a silent disaster-a gradual yet relentless process whose impacts accumulate over time,progressively degrading landscape integrity and disrupting ecological sustainability.Unlike catastrophic events with immediate visibility,soil erosion’s most devastating consequences often manifest decades later through diminished agricultural productivity,habitat fragmentation,and irreversible biodiversity loss.This study developed a scalable predictive framework employing Random Forest(RF)and Gradient Boosting Tree(GBT)machine learning models to assess and map soil erosion susceptibility across the region.A comprehensive geo-database was developed incorporating 11 erosion triggering factors:slope,elevation,rainfall,drainage density,topographic wetness index,normalized difference vegetation index,curvature,soil texture,land use,geology,and aspect.A total of 2,483 historical soil erosion locations were identified and randomly divided into two sets:70%for model building and 30%for validation purposes.The models revealed distinct spatial patterns of erosion risks,with GBT classifying 60.50%of the area as very low susceptibility,while RF identified 28.92%in this category.Notable differences emerged in high-risk zone identification,with GBT highlighting 7.42%and RF indicating 2.21%as very high erosion susceptibility areas.Both models demonstrated robust predictive capabilities,with GBT achieving 80.77%accuracy and 0.975 AUC,slightly outperforming RF’s 79.67%accuracy and 0.972 AUC.Analysis of predictor variables identified elevation,slope,rainfall and NDVI as the primary factors influencing erosion susceptibility,highlighting the complex interrelationship between geo-environmental factors and erosion processes.This research offers a strategic framework for targeted conservation and sustainable land management in the fragile Himalayan region,providing valuable insights to help policymakers implement effective soil erosion mitigation strategies and support long-term environmental sustainability.展开更多
To investigate the travel time prediction method of the freeway, a model based on the gradient boosting decision tree (GBDT) is proposed. Eleven variables (namely, travel time in current period T i , traffic flow in c...To investigate the travel time prediction method of the freeway, a model based on the gradient boosting decision tree (GBDT) is proposed. Eleven variables (namely, travel time in current period T i , traffic flow in current period Q i , speed in current period V i , density in current period K i , the number of vehicles in current period N i , occupancy in current period R i , traffic state parameter in current period X i , travel time in previous time period T i -1 , etc.) are selected to predict the travel time for 10 min ahead in the proposed model. Data obtained from VISSIM simulation is used to train and test the model. The results demonstrate that the prediction error of the GBDT model is smaller than those of the back propagation (BP) neural network model and the support vector machine (SVM) model. Travel time in current period T i is the most important variable among all variables in the GBDT model. The GBDT model can produce more accurate prediction results and mine the hidden nonlinear relationships deeply between variables and the predicted travel time.展开更多
Accurate prediction ofmonthly oil and gas production is essential for oil enterprises tomake reasonable production plans,avoid blind investment and realize sustainable development.Traditional oil well production trend...Accurate prediction ofmonthly oil and gas production is essential for oil enterprises tomake reasonable production plans,avoid blind investment and realize sustainable development.Traditional oil well production trend prediction methods are based on years of oil field production experience and expertise,and the application conditions are very demanding.With the rapid development of artificial intelligence technology,big data analysis methods are gradually applied in various sub-fields of the oil and gas reservoir development.Based on the data-driven artificial intelligence algorithmGradient BoostingDecision Tree(GBDT),this paper predicts the initial single-layer production by considering geological data,fluid PVT data and well data.The results show that the GBDT algorithm prediction model has great accuracy,significantly improving efficiency and strong universal applicability.The GBDTmethod trained in this paper can predict production,which is helpful for well site optimization,perforation layer optimization and engineering parameter optimization and has guiding significance for oilfield development.展开更多
The effects of the built environment factors on urban vitality have attracted wide attention in the urban planning fields in recent years,but few studies have considered the variables’relative importance and their no...The effects of the built environment factors on urban vitality have attracted wide attention in the urban planning fields in recent years,but few studies have considered the variables’relative importance and their nonlinear effects on urban vitality.Taking a Chinese metropolis—Hangzhou as a case study,this study applied the gradient boosting decision tree(GBDT)model to explore the nonlinear effects of the 5D factors of the urban built environment on urban social vitality and economic vitality and the importance of variables.The results show that the GBDT model has better goodness of fit than the traditional ordinary least squares(OLS)regression in the urban vitality models.The urban built environment plays an important role in affecting urban vitality,while built environment designs witness the most important effect.Specifically,the density of shopping facilities,medical facilities,and road networks are the most important factors affecting urban social vitality,while road network density,destination accessibility,and population density play the most important roles in affecting urban economic vitality.Finally,the urban built environment factors have nonlinear threshold effects on both urban economic and social vitality in Hangzhou,with differing nonlinear response patterns observed between social and economic dimensions.展开更多
The prediction of power grid faults based on meteorological factors is of great significance to reduce economic losses caused by power grid faults. However, the existing methods fail to effectively extract key feature...The prediction of power grid faults based on meteorological factors is of great significance to reduce economic losses caused by power grid faults. However, the existing methods fail to effectively extract key features and accurately predict fault types due to the complexity of meteorological factors and their nonlinear relationships. In response to these challenges, we propose the Feature-Enhanced XGBoost power grid fault prediction method (FE-XGBoost). Specifically, we first combine the gradient boosting decision tree and recursive feature elimination method to extract essential features from meteorological data. Then, we incorporate a piecewise linear chaotic map to enhance the optimization accuracy of the sparrow search algorithm. Finally, we construct an XGBoost-based model for the classification prediction of power grid meteorological faults and optimize the hyperparameters such as the optimal tree depth, optimal learning rate, and optimal number of iterations using an enhanced sparrow search algorithm. Experimental results demonstrate that our method outperforms the baseline models in predicting power grid faults accurately.展开更多
Understanding the impact of meteorological and topographical factors on snow cover fraction(SCF)is crucial for water resource management in the Qilian Mountains(QLM),China.However,there is still a lack of adequate qua...Understanding the impact of meteorological and topographical factors on snow cover fraction(SCF)is crucial for water resource management in the Qilian Mountains(QLM),China.However,there is still a lack of adequate quantitative analysis of the impact of these factors.This study investigated the spatiotemporal characteristics and trends of SCF in the QLM based on the cloud-removed Moderate Resolution Imaging Spectroradiometer(MODIS)SCF dataset during 2000-2021 and conducted a quantitative analysis of the drivers using a histogram-based gradient boosting regression tree(HGBRT)model.The results indicated that the monthly distribution of SCF exhibited a bimodal pattern.The SCF showed a pattern of higher values in the western regions and lower values in the eastern regions.Overall,the SCF showed a decreasing trend during 2000-2021.The decrease in SCF occurred at higher elevations,while an increase was observed at lower elevations.At the annual scale,the SCF showed a downward trend in the western regions affected by westerly(52.84%of the QLM).However,the opposite trend was observed in the eastern regions affected by monsoon(45.73%of the QLM).The SCF displayed broadly similar spatial patterns in autumn and winter,with a significant decrease in the western regions and a slight increase in the central and eastern regions.The effect of spring SCF on spring surface runoff was more pronounced than that of winter SCF.Furthermore,compared with meteorological factors,a variation of 46.53%in spring surface runoff can be attributed to changes in spring SCF.At the annual scale,temperature and relative humidity were the most important drivers of SCF change.An increase in temperature exceeding 0.04°C/a was observed to result in a decline in SCF,with a maximum decrease of 0.22%/a.An increase in relative humidity of more than 0.02%/a stabilized the rise in SCF(about 0.06%/a).The impacts of slope and aspect were found to be minimal.At the seasonal scale,the primary factors impacting SCF change varied.In spring,precipitation and wind speed emerged as the primary drivers.In autumn,precipitation and temperature were identified as the primary drivers.In winter,relative humidity and precipitation were the most important drivers.In contrast to the other seasons,slope exerted the strongest influence on SCF change in summer.This study facilitates a detailed quantitative description of SCF change in the QLM,enhancing the effectiveness of watershed water resource management and ecological conservation efforts in this region.展开更多
In order to improve the accuracy of target intent recognition,a recognition method based on XGBoost(eXtreme Gradient Boosting)decision tree is proposed.This paper adopts relevant data and program of python to calculat...In order to improve the accuracy of target intent recognition,a recognition method based on XGBoost(eXtreme Gradient Boosting)decision tree is proposed.This paper adopts relevant data and program of python to calculate the probability of tactical intention.Then the sequence intention probability is obtained by applying Dempster-Shafer rule of combination.To verify the accuracy of recognition results,we compare the experimental results of this paper with the results in the literatures.The experiment shows that the probability of tactical intention recognition through this method is improved,so this method is feasible.展开更多
It is easy for teenagers to view pornographic pictures on social networks. Many researchers have studied the detection of real pornographic pictures, but there are few studies on those that are artificial. In this wor...It is easy for teenagers to view pornographic pictures on social networks. Many researchers have studied the detection of real pornographic pictures, but there are few studies on those that are artificial. In this work, we studied how to detect artificial pornographic pictures, especially when they are on social networks. The whole detection process can be divided into two stages： feature selection and picture detection. In the feature selection stage, seven types of features that favour picture detection were selected. In the picture detection stage, three steps were included. 1） In order to alleviate the imbalance in the number of artificial pornographic pictures and normal ones, the training dataset of artificial pornographic pictures was expanded. Therefore, the features which were extracted from the training dataset can also be expanded too. 2） In order to reduce the time of feature extraction, a fast method which extracted features based on the proportionally scaled picture rather than the original one was proposed. 3） Three tree models were compared and a gradient boost decision tree （GBDT） was selected for the final picture detection. Three sets of experimental results show that the proposed method can achieve better recognition precision and drastically reduce the time cost of the method.展开更多
红绿灯位置是道路上行人和车辆的交会点,极大影响着道路结构和交通运行,在城市路网中起着重要的枢纽作用。针对目前红绿灯位置检测方法准确率不够高、覆盖面区域不完整等问题,提出了一种基于轨迹数据的交通灯位置检测方法。该方法基于聚...红绿灯位置是道路上行人和车辆的交会点,极大影响着道路结构和交通运行,在城市路网中起着重要的枢纽作用。针对目前红绿灯位置检测方法准确率不够高、覆盖面区域不完整等问题,提出了一种基于轨迹数据的交通灯位置检测方法。该方法基于聚类-合并-分类-合并的四级模型,首先从清理过的轨迹数据中提取隐含的车辆行驶特征,再采用具有噪声的基于密度的聚类(density-based spatial clustering of applications with noise,DBSCAN)方法得到转向和停驻两类聚类中心,对这两类聚类中心进行合并,获得红绿灯位置的候选位置;根据候选位置一定范围内的轨迹点提取该区域的车流行驶特征,然后采用梯度提升决策树(gradient boosting decision tree,GBDT)算法进行分类,最后将候选位置的正样本融合,以检测红绿灯位置。采用成都市浮动车GPS轨迹数据进行实验,检测结果的F1分数为0.947,效果优于常规的机器学习方法。实验结果表明,基于GPS轨迹数据,采用提出的四层模型能有效检测出红绿灯的位置,该模型可被用于城市大范围红绿灯位置信息的快速获取和更新。展开更多
Protein-protein interactions(PPIs)are of great importance to understand genetic mechanisms,delineate disease pathogenesis,and guide drug design.With the increase of PPI data and development of machine learning technol...Protein-protein interactions(PPIs)are of great importance to understand genetic mechanisms,delineate disease pathogenesis,and guide drug design.With the increase of PPI data and development of machine learning technologies,prediction and identification of PPIs have become a research hotspot in proteomics.In this study,we propose a new prediction pipeline for PPIs based on gradient tree boosting(GTB).First,the initial feature vector is extracted by fusing pseudo amino acid composition(Pse AAC),pseudo position-specific scoring matrix(Pse PSSM),reduced sequence and index-vectors(RSIV),and autocorrelation descriptor(AD).Second,to remove redundancy and noise,we employ L1-regularized logistic regression(L1-RLR)to select an optimal feature subset.Finally,GTB-PPI model is constructed.Five-fold cross-validation showed that GTB-PPI achieved the accuracies of 95.15% and 90.47% on Saccharomyces cerevisiae and Helicobacter pylori datasets,respectively.In addition,GTB-PPI could be applied to predict the independent test datasets for Caenorhabditis elegans,Escherichia coli,Homo sapiens,and Mus musculus,the one-core PPI network for CD9,and the crossover PPI network for the Wnt-related signaling pathways.The results show that GTB-PPI can significantly improve accuracy of PPI prediction.The code and datasets of GTB-PPI can be downloaded from https://github.com/QUST-AIBBDRC/GTB-PPI/.展开更多
Cable-stayed bridges have been widely used in high-speed railway infrastructure.The accurate determination of cable’s representative temperatures is vital during the intricate processes of design,construction,and mai...Cable-stayed bridges have been widely used in high-speed railway infrastructure.The accurate determination of cable’s representative temperatures is vital during the intricate processes of design,construction,and maintenance of cable-stayed bridges.However,the representative temperatures of stayed cables are not specified in the existing design codes.To address this issue,this study investigates the distribution of the cable temperature and determinates its representative temperature.First,an experimental investigation,spanning over a period of one year,was carried out near the bridge site to obtain the temperature data.According to the statistical analysis of the measured data,it reveals that the temperature distribution is generally uniform along the cable cross-section without significant temperature gradient.Then,based on the limited data,the Monte Carlo,the gradient boosted regression trees(GBRT),and univariate linear regression(ULR)methods are employed to predict the cable’s representative temperature throughout the service life.These methods effectively overcome the limitations of insufficient monitoring data and accurately predict the representative temperature of the cables.However,each method has its own advantages and limitations in terms of applicability and accuracy.A comprehensive evaluation of the performance of these methods is conducted,and practical recommendations are provided for their application.The proposed methods and representative temperatures provide a good basis for the operation and maintenance of in-service long-span cable-stayed bridges.展开更多
Sepsis poses a serious threat to health of children in pediatric intensive care unit.The mortality from pediatric sepsis can be effectively reduced through in-time diagnosis and therapeutic intervention.The bacillicul...Sepsis poses a serious threat to health of children in pediatric intensive care unit.The mortality from pediatric sepsis can be effectively reduced through in-time diagnosis and therapeutic intervention.The bacilliculture detection method is too time-consuming to receive timely treatment.In this research,we propose a new framework:a deep encoding network with cross features(CF-DEN)that enables accurate early detection of sepsis.Cross features are automatically constructed via the gradient boosting decision tree and distilled into the deep encoding network(DEN)we designed.The DEN is aimed at learning sufficiently effective representation from clinical test data.Each layer of the DEN fltrates the features involved in computation at current layer via attention mechanism and outputs the current prediction which is additive layer by layer to obtain the embedding feature at last layer.The framework takes the advantage of tree-based method and neural network method to extract effective representation from small clinical dataset and obtain accurate prediction in order to prompt patient to get timely treatment.We evaluate the performance of the framework on the dataset collected from Shanghai Children's Medical Center.Compared with common machine learning methods,our method achieves the increase on F1-score by 16.06%on the test set.展开更多
The stability of underground entry-type excavations will directly affect the working environment and the safety of staff.Empirical critical span graphs and traditional statistics learning methods can not meet the requ...The stability of underground entry-type excavations will directly affect the working environment and the safety of staff.Empirical critical span graphs and traditional statistics learning methods can not meet the requirements of high accuracy for stability assessment of entry-type excavations.Therefore,this study proposes a new prediction method based on machine learning to scientifically adjust the critical span graph.Accordingly,the particle swarm optimization(PSO)algorithm is used to optimize the core parameters of the gradient boosting decision tree(GBDT),abbreviated as PSO-GBDT.Moreover,the classification performance of eight other classifiers including GDBT,k-nearest neighbors(KNN),two kinds of support vector machines(SVM),Gaussian naive Bayes(GNB),logistic regression(LR)and linear discriminant analysis(LDA)are also applied to compare with the proposed model.Findings revealed that compared with the other eight models,the prediction performance of PSO-GBDT is undoubtedly the most reliable,and its classification accuracy is up to 0.93.Therefore,this model has great potential to provide a more scientific and accurate choice for the stability prediction of underground excavations.In addition,each classification model is used to predict the stability category of several grid points divided by the critical span graph,and the updated critical span graph of each model is discussed in combination with previous studies.The results show that the PSO-GBDT model has the advantages of being scientific,accurate and efficient in updating the critical span graph,and its output decision boundary has strict theoretical support,which can help mine operators make favorable economic decisions.展开更多
This paper presents a hybrid ensemble classifier combined synthetic minority oversampling technique(SMOTE),random search(RS)hyper-parameters optimization algorithm and gradient boosting tree(GBT)to achieve efficient a...This paper presents a hybrid ensemble classifier combined synthetic minority oversampling technique(SMOTE),random search(RS)hyper-parameters optimization algorithm and gradient boosting tree(GBT)to achieve efficient and accurate rock trace identification.A thirteen-dimensional database consisting of basic,vector,and discontinuity features is established from image samples.All data points are classified as either‘‘trace”or‘‘non-trace”to divide the ultimate results into candidate trace samples.It is found that the SMOTE technology can effectively improve classification performance by recommending an optimized imbalance ratio of 1:5 to 1:4.Then,sixteen classifiers generated from four basic machine learning(ML)models are applied for performance comparison.The results reveal that the proposed RS-SMOTE-GBT classifier outperforms the other fifteen hybrid ML algorithms for both trace and nontrace classifications.Finally,discussions on feature importance,generalization ability and classification error are conducted for the proposed classifier.The experimental results indicate that more critical features affecting the trace classification are primarily from the discontinuity features.Besides,cleaning up the sedimentary pumice and reducing the area of fractured rock contribute to improving the overall classification performance.The proposed method provides a new alternative approach for the identification of 3D rock trace.展开更多
Epilepsy is a very common worldwide neurological disorder that can affect a person’s quality of life at any age. People with epilepsy typically have recurrent seizures that can lead to injury or in some cases even de...Epilepsy is a very common worldwide neurological disorder that can affect a person’s quality of life at any age. People with epilepsy typically have recurrent seizures that can lead to injury or in some cases even death. Curing epilepsy requires risky surgery. If not, the patient may be subjected to a long drug treatment associated with lifestyle advice without guarantee of total recovery. However, regardless of the type of treatment performed, late treatment necessarily creates psychological instability in the patient. It is therefore important to be able to diagnose the disease as early as possible if we desire that the patient does not suffer from its consequences on their mental health. That is why the study aims to propose a model for detecting epilepsy in order to be able to identify it as early as possible, especially in newborns. The objective of the article is to propose a model for detecting epilepsy using data from electroencephalogram signals from 10 newborns. This model developed using the extra trees classifier technique offers the possibility of predicting epilepsy in infants with an accuracy of around 99.4%.展开更多
The compression modulus(Es)is one of the most significant soil parameters that affects the compressive deformation of geotechnical systems,such as foundations.However,it is difficult and sometime costly to obtain this...The compression modulus(Es)is one of the most significant soil parameters that affects the compressive deformation of geotechnical systems,such as foundations.However,it is difficult and sometime costly to obtain this parameter in engineering practice.In this study,we aimed to develop a non-parametric ensemble artificial intelligence(AI)approach to calculate the Es of soft clay in contrast to the traditional regression models proposed in previous studies.A gradient boosted regression tree(GBRT)algorithm was used to discern the non-linear pattern between input variables and the target response,while a genetic algorithm(GA)was adopted for tuning the GBRT model's hyper-parameters.The model was tested through 10-fold cross validation.A dataset of 221 samples from 65 engineering survey reports from Shanghai infrastructure projects was constructed to evaluate the accuracy of the new model5 s predictions.The mean squared error and correlation coefficient of the optimum GBRT model applied to the testing set were 0.13 and 0.91,respectively,indicating that the proposed machine learning(ML)model has great potential to improve the prediction of Es for soft clay.A comparison of the performance of empirical formulas and the proposed ML method for predicting foundation settlement indicated the rationality of the proposed ML model and its applicability to the compressive deformation of geotechnical systems.This model,however,cannot be directly applied to the prediction of Es in other sites due to its site specificity.This problem can be solved by retraining the model using local data.This study provides a useful reference for future multi-parameter prediction of soil behavior.展开更多
基金The National Natural Science Foundation of China(No.52361165658,52378318,52078459).
文摘To enhance the accuracy and efficiency of bridge damage identification,a novel data-driven damage identification method was proposed.First,convolutional autoencoder(CAE)was used to extract key features from the acceleration signal of the bridge structure through data reconstruction.The extreme gradient boosting tree(XGBoost)was then used to perform analysis on the feature data to achieve damage detection with high accuracy and high performance.The proposed method was applied in a numerical simulation study on a three-span continuous girder and further validated experimentally on a scaled model of a cable-stayed bridge.The numerical simulation results show that the identification errors remain within 2.9%for six single-damage cases and within 3.1%for four double-damage cases.The experimental validation results demonstrate that when the tension in a single cable of the cable-stayed bridge decreases by 20%,the method accurately identifies damage at different cable locations using only sensors installed on the main girder,achieving identification accuracies above 95.8%in all cases.The proposed method shows high identification accuracy and generalization ability across various damage scenarios.
文摘BACKGROUND Development of distant metastasis(DM)is a major concern during treatment of nasopharyngeal carcinoma(NPC).However,studies have demonstrated im-proved distant control and survival in patients with advanced NPC with the addition of chemotherapy to concomitant chemoradiotherapy.Therefore,precise prediction of metastasis in patients with NPC is crucial.AIM To develop a predictive model for metastasis in NPC using detailed magnetic resonance imaging(MRI)reports.METHODS This retrospective study included 792 patients with non-distant metastatic NPC.A total of 469 imaging variables were obtained from detailed MRI reports.Data were stratified and randomly split into training(50%)and testing sets.Gradient boosting tree(GBT)models were built and used to select variables for predicting DM.A full model comprising all variables and a reduced model with the top-five variables were built.Model performance was assessed by area under the curve(AUC).RESULTS Among the 792 patients,94 developed DM during follow-up.The number of metastatic cervical nodes(30.9%),tumor invasion in the posterior half of the nasal cavity(9.7%),two sides of the pharyngeal recess(6.2%),tubal torus(3.3%),and single side of the parapharyngeal space(2.7%)were the top-five contributors for predicting DM,based on their relative importance in GBT models.The testing AUC of the full model was 0.75(95%confidence interval[CI]:0.69-0.82).The testing AUC of the reduced model was 0.75(95%CI:0.68-0.82).For the whole dataset,the full(AUC=0.76,95%CI:0.72-0.82)and reduced models(AUC=0.76,95%CI:0.71-0.81)outperformed the tumor node-staging system(AUC=0.67,95%CI:0.61-0.73).CONCLUSION The GBT model outperformed the tumor node-staging system in predicting metastasis in NPC.The number of metastatic cervical nodes was identified as the principal contributing variable.
文摘This paper aims to design an optimizer followed by a Kawahara filter for optimal classification and prediction of employees’performance.The algorithm starts by processing data by a modified K-means technique as a hierarchical clustering method to quickly obtain the best features of employees to reach their best performance.The work of this paper consists of two parts.The first part is based on collecting data of employees to calculate and illustrate the performance of each employee.The second part is based on the classification and prediction techniques of the employee performance.This model is designed to help companies in their decisions about the employees’performance.The classification and prediction algorithms use the Gradient Boosting Tree classifier to classify and predict the features.Results of the paper give the percentage of employees which are expected to leave the company after predicting their performance for the coming years.Results also show that the Grasshopper Optimization,followed by“KF”with the Gradient Boosting Tree as classifier and predictor,is characterized by a high accuracy.The proposed algorithm is compared with other known techniques where our results are fund to be superior.
基金supported by the National Nat-ural Science Foundation of China(No.52203376)the National Key Research and Development Program of China(No.2023YFB3813200).
文摘Traditional 3Ni weathering steel cannot completely meet the requirements for offshore engineering development,resulting in the design of novel 3Ni steel with the addition of microalloy elements such as Mn or Nb for strength enhancement becoming a trend.The stress-assisted corrosion behavior of a novel designed high-strength 3Ni steel was investigated in the current study using the corrosion big data method.The information on the corrosion process was recorded using the galvanic corrosion current monitoring method.The gradi-ent boosting decision tree(GBDT)machine learning method was used to mine the corrosion mechanism,and the importance of the struc-ture factor was investigated.Field exposure tests were conducted to verify the calculated results using the GBDT method.Results indic-ated that the GBDT method can be effectively used to study the influence of structural factors on the corrosion process of 3Ni steel.Dif-ferent mechanisms for the addition of Mn and Cu to the stress-assisted corrosion of 3Ni steel suggested that Mn and Cu have no obvious effect on the corrosion rate of non-stressed 3Ni steel during the early stage of corrosion.When the corrosion reached a stable state,the in-crease in Mn element content increased the corrosion rate of 3Ni steel,while Cu reduced this rate.In the presence of stress,the increase in Mn element content and Cu addition can inhibit the corrosion process.The corrosion law of outdoor-exposed 3Ni steel is consistent with the law based on corrosion big data technology,verifying the reliability of the big data evaluation method and data prediction model selection.
文摘The Darjeeling Himalayan region,characterized by its complex topography and vulnerability to multiple environmental hazards,faces significant challenges including landslides,earthquakes,flash floods,and soil loss that critically threaten ecosystem stability.Among these challenges,soil erosion emerges as a silent disaster-a gradual yet relentless process whose impacts accumulate over time,progressively degrading landscape integrity and disrupting ecological sustainability.Unlike catastrophic events with immediate visibility,soil erosion’s most devastating consequences often manifest decades later through diminished agricultural productivity,habitat fragmentation,and irreversible biodiversity loss.This study developed a scalable predictive framework employing Random Forest(RF)and Gradient Boosting Tree(GBT)machine learning models to assess and map soil erosion susceptibility across the region.A comprehensive geo-database was developed incorporating 11 erosion triggering factors:slope,elevation,rainfall,drainage density,topographic wetness index,normalized difference vegetation index,curvature,soil texture,land use,geology,and aspect.A total of 2,483 historical soil erosion locations were identified and randomly divided into two sets:70%for model building and 30%for validation purposes.The models revealed distinct spatial patterns of erosion risks,with GBT classifying 60.50%of the area as very low susceptibility,while RF identified 28.92%in this category.Notable differences emerged in high-risk zone identification,with GBT highlighting 7.42%and RF indicating 2.21%as very high erosion susceptibility areas.Both models demonstrated robust predictive capabilities,with GBT achieving 80.77%accuracy and 0.975 AUC,slightly outperforming RF’s 79.67%accuracy and 0.972 AUC.Analysis of predictor variables identified elevation,slope,rainfall and NDVI as the primary factors influencing erosion susceptibility,highlighting the complex interrelationship between geo-environmental factors and erosion processes.This research offers a strategic framework for targeted conservation and sustainable land management in the fragile Himalayan region,providing valuable insights to help policymakers implement effective soil erosion mitigation strategies and support long-term environmental sustainability.
基金The National Natural Science Foundation of China(No.51478114,51778136)
文摘To investigate the travel time prediction method of the freeway, a model based on the gradient boosting decision tree (GBDT) is proposed. Eleven variables (namely, travel time in current period T i , traffic flow in current period Q i , speed in current period V i , density in current period K i , the number of vehicles in current period N i , occupancy in current period R i , traffic state parameter in current period X i , travel time in previous time period T i -1 , etc.) are selected to predict the travel time for 10 min ahead in the proposed model. Data obtained from VISSIM simulation is used to train and test the model. The results demonstrate that the prediction error of the GBDT model is smaller than those of the back propagation (BP) neural network model and the support vector machine (SVM) model. Travel time in current period T i is the most important variable among all variables in the GBDT model. The GBDT model can produce more accurate prediction results and mine the hidden nonlinear relationships deeply between variables and the predicted travel time.
文摘Accurate prediction ofmonthly oil and gas production is essential for oil enterprises tomake reasonable production plans,avoid blind investment and realize sustainable development.Traditional oil well production trend prediction methods are based on years of oil field production experience and expertise,and the application conditions are very demanding.With the rapid development of artificial intelligence technology,big data analysis methods are gradually applied in various sub-fields of the oil and gas reservoir development.Based on the data-driven artificial intelligence algorithmGradient BoostingDecision Tree(GBDT),this paper predicts the initial single-layer production by considering geological data,fluid PVT data and well data.The results show that the GBDT algorithm prediction model has great accuracy,significantly improving efficiency and strong universal applicability.The GBDTmethod trained in this paper can predict production,which is helpful for well site optimization,perforation layer optimization and engineering parameter optimization and has guiding significance for oilfield development.
基金National Social Science Foundation of China,No.20FJLB025National Natural Science Foundation of China,No.42471207,No.42471203Zhejiang Province Philosophy and Social Science Planning,Zhijiang Youth Special Project,24ZJQN118Y。
文摘The effects of the built environment factors on urban vitality have attracted wide attention in the urban planning fields in recent years,but few studies have considered the variables’relative importance and their nonlinear effects on urban vitality.Taking a Chinese metropolis—Hangzhou as a case study,this study applied the gradient boosting decision tree(GBDT)model to explore the nonlinear effects of the 5D factors of the urban built environment on urban social vitality and economic vitality and the importance of variables.The results show that the GBDT model has better goodness of fit than the traditional ordinary least squares(OLS)regression in the urban vitality models.The urban built environment plays an important role in affecting urban vitality,while built environment designs witness the most important effect.Specifically,the density of shopping facilities,medical facilities,and road networks are the most important factors affecting urban social vitality,while road network density,destination accessibility,and population density play the most important roles in affecting urban economic vitality.Finally,the urban built environment factors have nonlinear threshold effects on both urban economic and social vitality in Hangzhou,with differing nonlinear response patterns observed between social and economic dimensions.
基金supported by the Science and Technology Project of State Grid Jiangsu Electric Power Co.,Ltd.(Research on Power Meteorology Digitalization Application for Future Climate Scenarios and New Energy Operation Risks,J2023076).
文摘The prediction of power grid faults based on meteorological factors is of great significance to reduce economic losses caused by power grid faults. However, the existing methods fail to effectively extract key features and accurately predict fault types due to the complexity of meteorological factors and their nonlinear relationships. In response to these challenges, we propose the Feature-Enhanced XGBoost power grid fault prediction method (FE-XGBoost). Specifically, we first combine the gradient boosting decision tree and recursive feature elimination method to extract essential features from meteorological data. Then, we incorporate a piecewise linear chaotic map to enhance the optimization accuracy of the sparrow search algorithm. Finally, we construct an XGBoost-based model for the classification prediction of power grid meteorological faults and optimize the hyperparameters such as the optimal tree depth, optimal learning rate, and optimal number of iterations using an enhanced sparrow search algorithm. Experimental results demonstrate that our method outperforms the baseline models in predicting power grid faults accurately.
基金funded by the Key Research and Development Project for Ecological Civilization Construction in Gansu Province(24YFFA010)the Gansu Province Major Science and Technology Project(22ZD6FA005)+2 种基金the Natural Science Foundation of Gansu Province(24JRRA091)the Shanxi Province Basic Research Program(Free Exploration Category)Youth Project(202403021212316)the Science and Technology Innovation Program for Universities in Shanxi Province(2024L327)。
文摘Understanding the impact of meteorological and topographical factors on snow cover fraction(SCF)is crucial for water resource management in the Qilian Mountains(QLM),China.However,there is still a lack of adequate quantitative analysis of the impact of these factors.This study investigated the spatiotemporal characteristics and trends of SCF in the QLM based on the cloud-removed Moderate Resolution Imaging Spectroradiometer(MODIS)SCF dataset during 2000-2021 and conducted a quantitative analysis of the drivers using a histogram-based gradient boosting regression tree(HGBRT)model.The results indicated that the monthly distribution of SCF exhibited a bimodal pattern.The SCF showed a pattern of higher values in the western regions and lower values in the eastern regions.Overall,the SCF showed a decreasing trend during 2000-2021.The decrease in SCF occurred at higher elevations,while an increase was observed at lower elevations.At the annual scale,the SCF showed a downward trend in the western regions affected by westerly(52.84%of the QLM).However,the opposite trend was observed in the eastern regions affected by monsoon(45.73%of the QLM).The SCF displayed broadly similar spatial patterns in autumn and winter,with a significant decrease in the western regions and a slight increase in the central and eastern regions.The effect of spring SCF on spring surface runoff was more pronounced than that of winter SCF.Furthermore,compared with meteorological factors,a variation of 46.53%in spring surface runoff can be attributed to changes in spring SCF.At the annual scale,temperature and relative humidity were the most important drivers of SCF change.An increase in temperature exceeding 0.04°C/a was observed to result in a decline in SCF,with a maximum decrease of 0.22%/a.An increase in relative humidity of more than 0.02%/a stabilized the rise in SCF(about 0.06%/a).The impacts of slope and aspect were found to be minimal.At the seasonal scale,the primary factors impacting SCF change varied.In spring,precipitation and wind speed emerged as the primary drivers.In autumn,precipitation and temperature were identified as the primary drivers.In winter,relative humidity and precipitation were the most important drivers.In contrast to the other seasons,slope exerted the strongest influence on SCF change in summer.This study facilitates a detailed quantitative description of SCF change in the QLM,enhancing the effectiveness of watershed water resource management and ecological conservation efforts in this region.
文摘In order to improve the accuracy of target intent recognition,a recognition method based on XGBoost(eXtreme Gradient Boosting)decision tree is proposed.This paper adopts relevant data and program of python to calculate the probability of tactical intention.Then the sequence intention probability is obtained by applying Dempster-Shafer rule of combination.To verify the accuracy of recognition results,we compare the experimental results of this paper with the results in the literatures.The experiment shows that the probability of tactical intention recognition through this method is improved,so this method is feasible.
基金Projects(61573380,61303185) supported by the National Natural Science Foundation of ChinaProjects(2016M592450,2017M612585) supported by the China Postdoctoral Science FoundationProjects(2016JJ4119,2017JJ3416) supported by the Hunan Provincial Natural Science Foundation of China
文摘It is easy for teenagers to view pornographic pictures on social networks. Many researchers have studied the detection of real pornographic pictures, but there are few studies on those that are artificial. In this work, we studied how to detect artificial pornographic pictures, especially when they are on social networks. The whole detection process can be divided into two stages： feature selection and picture detection. In the feature selection stage, seven types of features that favour picture detection were selected. In the picture detection stage, three steps were included. 1） In order to alleviate the imbalance in the number of artificial pornographic pictures and normal ones, the training dataset of artificial pornographic pictures was expanded. Therefore, the features which were extracted from the training dataset can also be expanded too. 2） In order to reduce the time of feature extraction, a fast method which extracted features based on the proportionally scaled picture rather than the original one was proposed. 3） Three tree models were compared and a gradient boost decision tree （GBDT） was selected for the final picture detection. Three sets of experimental results show that the proposed method can achieve better recognition precision and drastically reduce the time cost of the method.
文摘红绿灯位置是道路上行人和车辆的交会点,极大影响着道路结构和交通运行,在城市路网中起着重要的枢纽作用。针对目前红绿灯位置检测方法准确率不够高、覆盖面区域不完整等问题,提出了一种基于轨迹数据的交通灯位置检测方法。该方法基于聚类-合并-分类-合并的四级模型,首先从清理过的轨迹数据中提取隐含的车辆行驶特征,再采用具有噪声的基于密度的聚类(density-based spatial clustering of applications with noise,DBSCAN)方法得到转向和停驻两类聚类中心,对这两类聚类中心进行合并,获得红绿灯位置的候选位置;根据候选位置一定范围内的轨迹点提取该区域的车流行驶特征,然后采用梯度提升决策树(gradient boosting decision tree,GBDT)算法进行分类,最后将候选位置的正样本融合,以检测红绿灯位置。采用成都市浮动车GPS轨迹数据进行实验,检测结果的F1分数为0.947,效果优于常规的机器学习方法。实验结果表明,基于GPS轨迹数据,采用提出的四层模型能有效检测出红绿灯的位置,该模型可被用于城市大范围红绿灯位置信息的快速获取和更新。
基金supported by the National Natural Science Foundation of China(Grant No.61863010)the Key Research and Development Program of Shandong Province of China(Grant No.2019GGX101001)the Natural Science Foundation of Shandong Province of China(Grant No.ZR2018MC007)。
文摘Protein-protein interactions(PPIs)are of great importance to understand genetic mechanisms,delineate disease pathogenesis,and guide drug design.With the increase of PPI data and development of machine learning technologies,prediction and identification of PPIs have become a research hotspot in proteomics.In this study,we propose a new prediction pipeline for PPIs based on gradient tree boosting(GTB).First,the initial feature vector is extracted by fusing pseudo amino acid composition(Pse AAC),pseudo position-specific scoring matrix(Pse PSSM),reduced sequence and index-vectors(RSIV),and autocorrelation descriptor(AD).Second,to remove redundancy and noise,we employ L1-regularized logistic regression(L1-RLR)to select an optimal feature subset.Finally,GTB-PPI model is constructed.Five-fold cross-validation showed that GTB-PPI achieved the accuracies of 95.15% and 90.47% on Saccharomyces cerevisiae and Helicobacter pylori datasets,respectively.In addition,GTB-PPI could be applied to predict the independent test datasets for Caenorhabditis elegans,Escherichia coli,Homo sapiens,and Mus musculus,the one-core PPI network for CD9,and the crossover PPI network for the Wnt-related signaling pathways.The results show that GTB-PPI can significantly improve accuracy of PPI prediction.The code and datasets of GTB-PPI can be downloaded from https://github.com/QUST-AIBBDRC/GTB-PPI/.
基金Project(2017G006-N)supported by the Project of Science and Technology Research and Development Program of China Railway Corporation。
文摘Cable-stayed bridges have been widely used in high-speed railway infrastructure.The accurate determination of cable’s representative temperatures is vital during the intricate processes of design,construction,and maintenance of cable-stayed bridges.However,the representative temperatures of stayed cables are not specified in the existing design codes.To address this issue,this study investigates the distribution of the cable temperature and determinates its representative temperature.First,an experimental investigation,spanning over a period of one year,was carried out near the bridge site to obtain the temperature data.According to the statistical analysis of the measured data,it reveals that the temperature distribution is generally uniform along the cable cross-section without significant temperature gradient.Then,based on the limited data,the Monte Carlo,the gradient boosted regression trees(GBRT),and univariate linear regression(ULR)methods are employed to predict the cable’s representative temperature throughout the service life.These methods effectively overcome the limitations of insufficient monitoring data and accurately predict the representative temperature of the cables.However,each method has its own advantages and limitations in terms of applicability and accuracy.A comprehensive evaluation of the performance of these methods is conducted,and practical recommendations are provided for their application.The proposed methods and representative temperatures provide a good basis for the operation and maintenance of in-service long-span cable-stayed bridges.
文摘Sepsis poses a serious threat to health of children in pediatric intensive care unit.The mortality from pediatric sepsis can be effectively reduced through in-time diagnosis and therapeutic intervention.The bacilliculture detection method is too time-consuming to receive timely treatment.In this research,we propose a new framework:a deep encoding network with cross features(CF-DEN)that enables accurate early detection of sepsis.Cross features are automatically constructed via the gradient boosting decision tree and distilled into the deep encoding network(DEN)we designed.The DEN is aimed at learning sufficiently effective representation from clinical test data.Each layer of the DEN fltrates the features involved in computation at current layer via attention mechanism and outputs the current prediction which is additive layer by layer to obtain the embedding feature at last layer.The framework takes the advantage of tree-based method and neural network method to extract effective representation from small clinical dataset and obtain accurate prediction in order to prompt patient to get timely treatment.We evaluate the performance of the framework on the dataset collected from Shanghai Children's Medical Center.Compared with common machine learning methods,our method achieves the increase on F1-score by 16.06%on the test set.
基金the National Science Foundation of China(Grant No.42177164)the Distinguished Youth Science Foundation of Hunan Province of China(Grant No.2022JJ10073)the Innovation-Driven Project of Central South University(Grant No.2020CX040).
文摘The stability of underground entry-type excavations will directly affect the working environment and the safety of staff.Empirical critical span graphs and traditional statistics learning methods can not meet the requirements of high accuracy for stability assessment of entry-type excavations.Therefore,this study proposes a new prediction method based on machine learning to scientifically adjust the critical span graph.Accordingly,the particle swarm optimization(PSO)algorithm is used to optimize the core parameters of the gradient boosting decision tree(GBDT),abbreviated as PSO-GBDT.Moreover,the classification performance of eight other classifiers including GDBT,k-nearest neighbors(KNN),two kinds of support vector machines(SVM),Gaussian naive Bayes(GNB),logistic regression(LR)and linear discriminant analysis(LDA)are also applied to compare with the proposed model.Findings revealed that compared with the other eight models,the prediction performance of PSO-GBDT is undoubtedly the most reliable,and its classification accuracy is up to 0.93.Therefore,this model has great potential to provide a more scientific and accurate choice for the stability prediction of underground excavations.In addition,each classification model is used to predict the stability category of several grid points divided by the critical span graph,and the updated critical span graph of each model is discussed in combination with previous studies.The results show that the PSO-GBDT model has the advantages of being scientific,accurate and efficient in updating the critical span graph,and its output decision boundary has strict theoretical support,which can help mine operators make favorable economic decisions.
基金supported by Key innovation team program of innovation talents promotion plan by MOST of China(No.2016RA4059)Natural Science Foundation Committee Program of China(No.51778474)Science and Technology Project of Yunnan Provincial Transportation Department(No.25 of 2018)。
文摘This paper presents a hybrid ensemble classifier combined synthetic minority oversampling technique(SMOTE),random search(RS)hyper-parameters optimization algorithm and gradient boosting tree(GBT)to achieve efficient and accurate rock trace identification.A thirteen-dimensional database consisting of basic,vector,and discontinuity features is established from image samples.All data points are classified as either‘‘trace”or‘‘non-trace”to divide the ultimate results into candidate trace samples.It is found that the SMOTE technology can effectively improve classification performance by recommending an optimized imbalance ratio of 1:5 to 1:4.Then,sixteen classifiers generated from four basic machine learning(ML)models are applied for performance comparison.The results reveal that the proposed RS-SMOTE-GBT classifier outperforms the other fifteen hybrid ML algorithms for both trace and nontrace classifications.Finally,discussions on feature importance,generalization ability and classification error are conducted for the proposed classifier.The experimental results indicate that more critical features affecting the trace classification are primarily from the discontinuity features.Besides,cleaning up the sedimentary pumice and reducing the area of fractured rock contribute to improving the overall classification performance.The proposed method provides a new alternative approach for the identification of 3D rock trace.
文摘Epilepsy is a very common worldwide neurological disorder that can affect a person’s quality of life at any age. People with epilepsy typically have recurrent seizures that can lead to injury or in some cases even death. Curing epilepsy requires risky surgery. If not, the patient may be subjected to a long drug treatment associated with lifestyle advice without guarantee of total recovery. However, regardless of the type of treatment performed, late treatment necessarily creates psychological instability in the patient. It is therefore important to be able to diagnose the disease as early as possible if we desire that the patient does not suffer from its consequences on their mental health. That is why the study aims to propose a model for detecting epilepsy in order to be able to identify it as early as possible, especially in newborns. The objective of the article is to propose a model for detecting epilepsy using data from electroencephalogram signals from 10 newborns. This model developed using the extra trees classifier technique offers the possibility of predicting epilepsy in infants with an accuracy of around 99.4%.
基金the National Natural Science Foundation of China(Nos.51608380 and 51538009)the Key Innovation Team Program of the Innovation Talents Promotion Plan by Ministry of Science and Technology of China(No.2016RA4059)the Specific Consultant Research Project of Shanghai Tunnel Engineering Company Ltd.(No.STEC/KJB/XMGL/0130),China。
文摘The compression modulus(Es)is one of the most significant soil parameters that affects the compressive deformation of geotechnical systems,such as foundations.However,it is difficult and sometime costly to obtain this parameter in engineering practice.In this study,we aimed to develop a non-parametric ensemble artificial intelligence(AI)approach to calculate the Es of soft clay in contrast to the traditional regression models proposed in previous studies.A gradient boosted regression tree(GBRT)algorithm was used to discern the non-linear pattern between input variables and the target response,while a genetic algorithm(GA)was adopted for tuning the GBRT model's hyper-parameters.The model was tested through 10-fold cross validation.A dataset of 221 samples from 65 engineering survey reports from Shanghai infrastructure projects was constructed to evaluate the accuracy of the new model5 s predictions.The mean squared error and correlation coefficient of the optimum GBRT model applied to the testing set were 0.13 and 0.91,respectively,indicating that the proposed machine learning(ML)model has great potential to improve the prediction of Es for soft clay.A comparison of the performance of empirical formulas and the proposed ML method for predicting foundation settlement indicated the rationality of the proposed ML model and its applicability to the compressive deformation of geotechnical systems.This model,however,cannot be directly applied to the prediction of Es in other sites due to its site specificity.This problem can be solved by retraining the model using local data.This study provides a useful reference for future multi-parameter prediction of soil behavior.
