Extremely Randomized Trees Research Articles

Using multivariate adaptive regression splines and extremely randomized trees algorithms to predict dust events frequency around an international wetland and prioritize its drivers.

This study aimed to evaluate the performance of multivariate adaptive regression splines (MARS) and extremely randomized trees (ERT) models for predicting the internal and external dust events frequencies (DEF) across the northeastern and southwestern regions of the Gavkhouni International Wetland. These models were also evaluated to model the internal DEF (IDEF) across the northwestern, southeastern, northern, and western regions around the wetland. Furthermore, the main factors controlling DEF and IDEF were identified based on the importance value (IV) of predictors in the best model. The results showed that the ERT model increased the prediction accuracies by an average of 40%, compared to the MARS model. According to the IV obtained from the ERT model, aerosol optical depth (IV = 0.28), wetland discharge (IV = 0.25), near-surface wind speed (IV = 0.08), and erosive winds frequency (IV = 0.07) were identified as the most important factors controlling DEF across the northeastern and southwestern regions of the wetland. However, the erosive wind speed was detected as the major factor affecting the IDEF in the northern (IV = 0.16), western (IV = 0.18), and southeastern (IV = 0.65) regions of study wetland. It was also found that vapor pressure with IV of 0.32 had the greatest effect on IDEF variations across the northwestern region of the wetland. Overall, the results demonstrate the effectiveness of the ERT model in modeling the factors affecting DEF and IDEF, and the results may be used to mitigate dust events hazards around the Gavkhouni Wetland, in central Iran.

Cost-Sensitive Extremely Randomized Trees Algorithm for Online Fault Detection of Wind Turbine Generators

The number of normal samples of wind turbine generators is much larger than the number of fault samples. To solve the problem of imbalanced classification in wind turbine generator fault detection, a cost-sensitive extremely randomized trees (CS-ERT) algorithm is proposed in this paper, in which the cost-sensitive learning method is introduced into an extremely randomized trees (ERT) algorithm. Based on the classification misclassification cost and class distribution, the misclassification cost gain (MCG) is proposed as the score measure of the CS-ERT model growth process to improve the classification accuracy of minority classes. The Hilbert-Schmidt independence criterion lasso (HSICLasso) feature selection method is used to select strongly correlated non-redundant features of doubly-fed wind turbine generators. The effectiveness of the method was verified by experiments on four different failure datasets of wind turbine generators. The experiment results show that average missing detection rate, average misclassification cost and gMean of the improved algorithm better than those of the ERT algorithm. In addition, compared with the CSForest, AdaCost and MetaCost methods, the proposed method has better real-time fault detection performance.

ET-score: Improving Protein-ligand Binding Affinity Prediction Based on Distance-weighted Interatomic Contact Features Using Extremely Randomized Trees Algorithm.

The molecular docking simulation is a key computational tool in modern drug discovery research that its predictive performance strongly depends on the employed scoring functions. Many recent studies have shown that the application of machine learning algorithms in the development of scoring functions has led to a significant improvement in docking performance. In this work, we introduce a new machine learning (ML) based scoring function called ET-Score, which employs the distance-weighted interatomic contacts between atom type pairs of the ligand and the protein for featurizing protein-ligand complexes and Extremely Randomized Trees algorithm for the training process. The performance of ET-Score is compared with some successful ML-based scoring functions and several popular classical scoring functions on the PDBbind 2016v core set. It is shown that our ET-Score model (with Pearson's correlation of 0.827 and RMSE of 1.332) achieves very good performance in comparison with most of the ML-based scoring functions and all classical scoring functions despite its extremely low computational cost. ET-Score's codes are freely available on the web at https://github.com/miladrayka/ET_Score.

Leveraging fine-grained mobile data for churn detection through Essence Random Forest

The rise of unstructured data leads to unprecedented opportunities for marketing applications along with new methodological challenges to leverage such data. In particular, redundancy among the features extracted from this data deserves special attention as it might prevent current methods to benefit from it. In this study, we propose to investigate the value of multiple fine-grained data sources i.e. websurfing, use of applications and geospatial mobility for churn detection within telephone companies. This value is analysed both in substitution and in complement to the value of the well-known communication network. What is more, we also suggest an adaptation of the Random Forest algorithm called Essence Random Forest designed to better address redundancy among extracted features. Analysing fine-grained data of a telephone company, we first find that geo-spatial mobility data might be a good long term alternative to the classical communication network that might become obsolete due to the competition with digital communications. Then, we show that, on the short term, these alternative fine-grained data might complement the communication network for an improved churn detection. In addition, compared to Random Forest and Extremely Randomized Trees, Essence Random Forest better leverages the value of unstructured data by offering an enhanced churn detection regardless of the addressed perspective i.e. substitution or complement. Finally, Essence Random Forest converges faster to stable results which is a salient property in a resource constrained environment.

A Novel Hybrid Fuel Consumption Prediction Model for Ocean-Going Container Ships Based on Sensor Data

Accurate, reliable, and real-time prediction of ship fuel consumption is the basis and premise of the development of fuel optimization; however, ship fuel consumption data mainly come from noon reports, and many current modeling methods have been based on a single model; therefore they have low accuracy and robustness. In this study, we propose a novel hybrid fuel consumption prediction model based on sensor data collected from an ocean-going container ship. First, a data processing method is proposed to clean the collected data. Secondly, the Bayesian optimization method of hyperparameters is used to reasonably set the hyperparameter values of the model. Finally, a hybrid fuel consumption prediction model is established by integrating extremely randomized tree (ET), random forest (RF), Xgboost (XGB) and multiple linear regression (MLR) methods. The experimental results show that data cleaning, the size of the dataset, marine environmental factors, and hyperparameter optimization can all affect the accuracy of the model, and the proposed hybrid model provides better predictive performance (higher accuracy) and greater robustness (smaller standard deviation) as compared with a single model. The proposed hybrid model should play a significant role in ship fuel consumption real-time monitoring, fault diagnosis, energy saving and emission reduction, etc.

BCD-WERT: a novel approach for breast cancer detection using whale optimization based efficient features and extremely randomized tree algorithm.

Breast cancer is one of the leading causes of death in the current age. It often results in subpar living conditions for a patient as they have to go through expensive and painful treatments to fight this cancer. One in eight women all over the world is affected by this disease. Almost half a million women annually do not survive this fight and die from this disease. Machine learning algorithms have proven to outperform all existing solutions for the prediction of breast cancer using models built on the previously available data. In this paper, a novel approach named BCD-WERT is proposed that utilizes the Extremely Randomized Tree and Whale Optimization Algorithm (WOA) for efficient feature selection and classification. WOA reduces the dimensionality of the dataset and extracts the relevant features for accurate classification. Experimental results on state-of-the-art comprehensive dataset demonstrated improved performance in comparison with eight other machine learning algorithms: Support Vector Machine (SVM), Random Forest, Kernel Support Vector Machine, Decision Tree, Logistic Regression, Stochastic Gradient Descent, Gaussian Naive Bayes and k-Nearest Neighbor. BCD-WERT outperformed all with the highest accuracy rate of 99.30% followed by SVM achieving 98.60% accuracy. Experimental results also reveal the effectiveness of feature selection techniques in improving prediction accuracy.

Nondestructive freshness evaluation of intact prawns (Fenneropenaeus chinensis) using line-scan spatially offset Raman spectroscopy

Prawns are highly popular with consumers but present many technical difficulties for the evaluation of their internal quality when intact (in-shell prawns). This study proposed a nondestructive method to assess the internal quality of intact prawns (Fenneropenaeus chinensis) using spatially offset Raman spectroscopy (SORS) technique combined with data modeling analysis. This technique holds promise due to the capability of SORS to obtain chemical information nondestructively from below the surface of a sample material. Raman scattering image data for 100 fresh prawns (approximately 15 g each) were collected using a line-scan Raman imaging system over the course of seven days with 24 h measurement intervals. Measurement anomalies due to physical prawn irregularities were eliminated using a peak identification method. Twenty feature bands selected by Random Forest (RF) method were input to Partial Least Squares Regression (PLSR), Support Vector Regression (SVR), and Extremely Randomized Tree (ET) models to predict the freshness of prawns during the storage time. The prediction model based on SORS enhanced data and combining RF feature band selection with SVR demonstrated the best performance, with RMSEP, R2, and RPD values of 0.71, 0.88, and 2.63, respectively. This rapid and nondestructive method for quality evaluation may be feasible as a practical means of assessing internal quality of materials that demonstrate surface interference, such as in-shell prawns.

Balancing accuracy and diversity in ensemble learning using a two-phase artificial bee colony approach

In ensemble learning, it is necessary to build a balancing mechanism to balance the accuracy of individual learners with the diversity between individual learners to achieve excellent ensemble learning performance. In previous studies, diversity was regarded only as a regularization term, which does not sufficiently indicate that diversity should implicitly be treated as an accuracy factor. In this study, an ensemble learning approach based on balanced accuracy and diversity (ELBAD) that uses a two-phase artificial bee colony (ABC) algorithm is proposed to balance the accuracy and diversity of ensemble learners. In the first phase, the ABC algorithm is used to generate an ensemble classifier with appropriate diversity. In the second phase, the ABC algorithm is used to generate a weighted ensemble classifier. The ELBAD ensemble learning algorithm is significantly superior to other state-of-the-art popular ensemble learning algorithms, including AdaBoost, Bagging, Decorate, extremely randomized trees (ET), gradient boosting decision tree (GBDT), random forest (RF), and rotation forest (RoF) on 30 UCI datasets. In addition, this study proposes a systematic parameter tuning procedure for the ELBAD algorithm that reduces the time required to generate an ensemble classifier.

Towards TBM Automation: On-The-Fly Characterization and Classification of Ground Conditions Ahead of a TBM Using Data-Driven Approach

Pre-tunneling exploration for rock mass classification is a common practice in tunneling projects. This study proposes a data-driven approach that allows for rock mass classification. Two machine learning (ML) classification models, namely random forest (RF) and extremely randomized tree (ERT), are employed to classify the rock mass conditions encountered in the Pahang-Selangor Raw Water Tunnel in Malaysia using tunnel boring machine (TBM) operating parameters. Due to imbalance of rock classes distribution, an oversampling technique was used to obtain a balanced training dataset for unbiased learning of the ML models. A five-fold cross-validation approach was used to tune the model hyperparameters and validation-set approach was used for the model evaluation. ERT achieved an overall accuracy of 95%, while RF achieved 94% accuracy, in rightly classifying rock mass conditions. The result shows that the proposed approach has the potential to identify and correctly classify ground conditions of a TBM, which allows for early problem detection and on-the-fly support system selection based on the identified ground condition. This study, which is part of an ongoing effort towards developing reliable models that could be incorporated into TBMs, shows the potential of data-driven approaches for on-the-fly classification of ground conditions ahead of a TBM and could allow for the early detection of potential construction problems.

The miniJPAS survey: star-galaxy classification using machine learning

Context.Future astrophysical surveys such as J-PAS will produce very large datasets, the so-called “big data”, which will require the deployment of accurate and efficient machine-learning (ML) methods. In this work, we analyze the miniJPAS survey, which observed about ∼1 deg2of the AEGIS field with 56 narrow-band filters and 4ugribroad-band filters. The miniJPAS primary catalog contains approximately 64 000 objects in therdetection band (magAB ≲ 24), with forced-photometry in all other filters.Aims.We discuss the classification of miniJPAS sources into extended (galaxies) and point-like (e.g., stars) objects, which is a step required for the subsequent scientific analyses. We aim at developing an ML classifier that is complementary to traditional tools that are based on explicit modeling. In particular, our goal is to release a value-added catalog with our best classification.Methods.In order to train and test our classifiers, we cross-matched the miniJPAS dataset with SDSS and HSC-SSP data, whose classification is trustworthy within the intervals 15 ≤ r ≤ 20 and 18.5 ≤ r ≤ 23.5, respectively. We trained and tested six different ML algorithms on the two cross-matched catalogs: K-nearest neighbors, decision trees, random forest (RF), artificial neural networks, extremely randomized trees (ERT), and an ensemble classifier. This last is a hybrid algorithm that combines artificial neural networks and RF with the J-PAS stellar and galactic loci classifier. As input for the ML algorithms we used the magnitudes from the 60 filters together with their errors, with and without the morphological parameters. We also used the mean point spread function in therdetection band for each pointing.Results.We find that the RF and ERT algorithms perform best in all scenarios. When the full magnitude range of 15 ≤ r ≤ 23.5 is analyzed, we find an area under the curve AUC = 0.957 with RF when photometric information alone is used, and AUC = 0.986 with ERT when photometric and morphological information is used together. When morphological parameters are used, the full width at half maximum is the most important feature. When photometric information is used alone, we observe that broad bands are not necessarily more important than narrow bands, and errors (the width of the distribution) are as important as the measurements (central value of the distribution). In other words, it is apparently important to fully characterize the measurement.Conclusions.ML algorithms can compete with traditional star and galaxy classifiers; they outperform the latter at fainter magnitudes (r ≳ 21). We use our best classifiers, with and without morphology, in order to produce a value-added catalog.

Real-Time Measurement of Seam Strength Using Optical Spectroscopy for Al–Li Alloy in Laser Beam Welding

This article investigates a real-time nondestructive measurement method of the seam strength using the optical spectroscopy and ensemble learning in laser beam welding. First, a linear positive relationship between the seam strength and acquired optical spectrum signal was established. Then, a new feature reduction method was proposed to extract the independent feature set with low correlation and high diversity. Finally, a hybrid classification model is proposed based on fast independent component analysis (ICA) and extremely randomized trees (ET). The model was thoroughly verified with various welding experiments and careful comparison with decision tree (DT), extreme gradient boosting DT (XGBoost), and support vector machine (SVM). More importantly, driven by the monitoring data and random forest with fast independent component analysis (RFICA)-ET model, it was found that Al I at 669.8673 nm and Ar I at 610.5635 nm were the two key elements in the dynamic plasma for accurate seam strength measurement.

Can Machine Learning Unlock the Continuous Alpha? Empirical Study Based on China A-Share Market

With the development of fintech and artificial intelligence, machine learning algorithms are widely used in quantitative investment. Based on the listed companies in China A-share market from February 2005 to July 2020, quantitative stock selection models with machine learning algorithms are established to obtain continuous alpha returns. The results show that machine learning algorithms can effectively identify the relationship between factors and returns and then improve the performance of the quantitative stock selection model. China A-share market is a weak-form efficient market. By mining the factors that are not fully digested by the market, continuous alpha returns can be obtained. The ensemble algorithms represented by the extremely randomized tree (ET) and light gradient boosting machine (LGBM) perform best in stock market prediction.

Improving the Performance of Machine Learning-Based Network Intrusion Detection Systems on the UNSW-NB15 Dataset.

Networks are exposed to an increasing number of cyberattacks due to their vulnerabilities. So, cybersecurity strives to make networks as safe as possible, by introducing defense systems to detect any suspicious activities. However, firewalls and classical intrusion detection systems (IDSs) suffer from continuous updating of their defined databases to detect threats. The new directions of the IDSs aim to leverage the machine learning models to design more robust systems with higher detection rates and lower false alarm rates. This research presents a novel network IDS, which plays an important role in network security and faces the current cyberattacks on networks using the UNSW-NB15 dataset benchmark. Our proposed system is a dynamically scalable multiclass machine learning-based network IDS. It consists of several stages based on supervised machine learning. It starts with the Synthetic Minority Oversampling Technique (SMOTE) method to solve the imbalanced classes problem in the dataset and then selects the important features for each class existing in the dataset by the Gini Impurity criterion using the Extremely Randomized Trees Classifier (Extra Trees Classifier). After that, a pretrained extreme learning machine (ELM) model is responsible for detecting the attacks separately, “One-Versus-All” as a binary classifier for each of them. Finally, the ELM classifier outputs become the inputs to a fully connected layer in order to learn from all their combinations, followed by a logistic regression layer to make soft decisions for all classes. Results show that our proposed system performs better than related works in terms of accuracy, false alarm rate, Receiver Operating Characteristic (ROC), and Precision-Recall Curves (PRCs).

Flash Flood Susceptibility Modeling Using New Approaches of Hybrid and Ensemble Tree-Based Machine Learning Algorithms

Flash flooding is considered one of the most dynamic natural disasters for which measures need to be taken to minimize economic damages, adverse effects, and consequences by mapping flood susceptibility. Identifying areas prone to flash flooding is a crucial step in flash flood hazard management. In the present study, the Kalvan watershed in Markazi Province, Iran, was chosen to evaluate the flash flood susceptibility modeling. Thus, to detect flash flood-prone zones in this study area, five machine learning (ML) algorithms were tested. These included boosted regression tree (BRT), random forest (RF), parallel random forest (PRF), regularized random forest (RRF), and extremely randomized trees (ERT). Fifteen climatic and geo-environmental variables were used as inputs of the flash flood susceptibility models. The results showed that ERT was the most optimal model with an area under curve (AUC) value of 0.82. The rest of the models’ AUC values, i.e., RRF, PRF, RF, and BRT, were 0.80, 0.79, 0.78, and 0.75, respectively. In the ERT model, the areal coverage for very high to moderate flash flood susceptible area was 582.56 km2 (28.33%), and the rest of the portion was associated with very low to low susceptibility zones. It is concluded that topographical and hydrological parameters, e.g., altitude, slope, rainfall, and the river’s distance, were the most effective parameters. The results of this study will play a vital role in the planning and implementation of flood mitigation strategies in the region.

A New Long-Term Downward Surface Solar Radiation Dataset over China from 1958 to 2015.

Downward surface solar radiation (Rs) plays a dominant role in determining the climate and environment on the Earth. However, the densely distributed ground observations of Rs are usually insufficient to meet the increasing demand of the climate diagnosis and analysis well, so it is essential to build a long-term accurate Rs dataset. The extremely randomized trees (ERT) algorithm was used to generate Rs using routine meteorological observations (2000–2015) from the Climate Data Center of the Chinese Meteorological Administration (CDC/CMA). The estimated Rs values were validated against ground measurements at the national scale with an overall correlation coefficient value of 0.97, a mean bias of 0.04 Wm−2, a root-mean-square-error value of 23.12 Wm−2, and a mean relative error of 9.81%. It indicates that the estimated Rs from the ERT-based model is reasonably accurate. Moreover, the ERT-based model was used to generate a new daily Rs dataset at 756 CDC/CMA stations from 1958 to 2015. The long-term variation trends of Rs at 454 stations covering 46 consecutive years (1970–2015) were also analyzed. The Rs in China showed a significant decline trend (−1.1 Wm−2 per decade) during 1970–2015. A decreasing trend (−2.8 Wm−2 per decade) in Rs during 1970–1992 was observed, followed by a recovery trend (0.23 Wm−2 per decade) during 1992–2015. The recovery trends at individual stations were found at 233 out of 454 stations during 1970–2015, which were mainly located in southern and northern China. The new Rs dataset would substantially provide basic data for the related studies in agriculture, ecology, and meteorology.

Strong classification system for wear identification on milling processes using computer vision and ensemble learning

Metallic pieces are typically machined by continuous tool passes, which usually causes regular patterns in the form of straight edges in the surface of the pieces. An irregular pattern on the surface of the piece implies a decrease of the quality of the machined piece. In this paper, we propose an acquisition system and a machine-vision based method to describe the texture of the inner and outer surfaces of machined pieces with cylindrical holes. In order to capture images of the hole surface, we used a microscope camera connected to a rigid industrial boroscope. Considering the extracted texture descriptors, a significant correlation is shown. Consequently, the feature vector is reduced and then classified by several algorithms using an exhaustive grid search strategy with 10-fold cross validation. Best results are achieved with the Extremely Randomized Trees classifier with a mean test score on the hold out set of 92.98%, what improves previous research and meets the requirements of the field.

Open-Circuit Fault Diagnosis of Three-Phase PWM Rectifier Using Beetle Antennae Search Algorithm Optimized Deep Belief Network

Effective open-circuit fault diagnosis for a two-level three-phase pulse-width modulating (PWM) rectifier can reduce the failure rate and prevent unscheduled shutdown. Nevertheless, traditional signal-based feature extraction methods show poor distinguishability for insufficient fault features. Shallow learning diagnosis models are prone to fall into local extremum, slow convergence speed, and overfitting. In this paper, a novel fault diagnosis strategy based on modified ensemble empirical mode decomposition (MEEMD) and the beetle antennae search (BAS) algorithm optimized deep belief network (DBN) is proposed to cope with these problems. Initially, MEEMD is applied to extract useful fault features from each intrinsic mode function (IMF) component. Meanwhile, to remove features with redundancy and interference, fault features are selected by calculating the importance of each feature based on the extremely randomized trees (ERT) algorithm, and the dimension of fault feature vectors is reduced by principal component analysis. Additionally, the DBN stacked with two layers of a restricted Boltzmann machine (RBM) is selected as the classifier, and the BAS algorithm is used as the optimizer to determine the optimal number of units in the hidden layers of the DBN. The proposed method combined with feature extraction, feature selection, optimization, and fault classification algorithms significantly improves the diagnosis accuracy.

Analysis of the Contribution Rate of the Influencing Factors to Land Subsidence in the Eastern Beijing Plain, China Based on Extremely Randomized Trees (ERT) Method

As a common geological hazard, land subsidence is widely distributed in the Eastern Beijing Plain. The pattern of evolution of this geological phenomenon is controlled by many factors, including groundwater level change in different aquifers, compressible layers of different thicknesses, and static and dynamic loads. First, based on the small baseline subset Interferometric Synthetic Aperture Radar (SBAS-InSAR) technique, we employed 47 ENVISAT ASAR images and 48 RADARSAT-2 images to acquire the ground deformation of the Beijing Plain from June 2003 to November 2015 and then validated the results using leveling benchmark monitoring. Second, we innovatively calculated additional stress to obtain static and dynamic load information. Finally, we evaluated the contribution rate of the influencing factors to land subsidence by using the Spearman’s rank correlation coefficient (SRCC) and extremely randomized trees (ERT) machine learning methods. The SBAS-InSAR outcomes revealed that the maximum deformation rate was 110.7 mm/year from 2003 to 2010 and 144.4 mm/year from 2010 to 2015. The SBAS-InSAR results agreed well with the leveling benchmark monitoring results; the correlation coefficients were 0.97 and 0.96 during the 2003–2010 and 2013–2015 periods, respectively. The contribution rate of the second confined aquifer to the cumulative land subsidence was 49.3% from 2003 to 2010, accounting for the largest proportion; however, its contribution rate decreased to 23.4% from 2010 to 2015. The contribution rate of the third confined aquifer to the cumulative land subsidence increased from 2003 to 2015. Although the contribution of additional stress engendered from static and dynamic loads to the cumulative land subsidence was slight, it had a significant effect on the uneven land subsidence, with a contribution rate of 33.8% from 2003 to 2010 and 23.1% from 2010 to 2015. These findings provide scientific support for mitigating hazards associated with land subsidence.

A Classification Framework using a Diverse Intensified Strawberry Optimized Neural Network (DISON) for Clinical Decision-making

A novel classification framework for clinical decision making that uses an Extremely Randomized Tree (ERT) based feature selection and a Diverse Intensified Strawberry Optimized Neural network (DISON) is proposed. DISON is a Feed Forward Artificial Neural Network where the optimization of weights and bias is done using a two phase training strategy. Two algorithms namely Strawberry Plant Optimization (SPO) algorithm and Gradient-descent Back-propagation algorithm are used sequentially to identify the optimum weights and bias. The novel two phase training method and the stochastic duplicate-elimination strategy of SPO helps in addressing the issue of local optima associated with conventional neural networks. The relevant attributes are selected based on the feature importance values computed using an ERT classifier.Vertebral Column, Pima Indian diabetes (PID), Cleveland Heart disease (CHD) and Statlog Heart disease (SHD) datasets from the University of California Irvine machine learning repository are used for experimentation. The framework has achieved an accuracy of 87.17% for Vertebral Column, 90.92% for PID, 93.67% for CHD and 94.5% for SHD. The classifier performance has been compared with existing works and is found to be competitive in terms of accuracy, sensitivity and specificity. Wilcoxon test confirms the statistical superiority of the proposed method.

Ensemble machine learning approaches for webshell detection in Internet of things environments

AbstractThe Internet of things (IoT), made up of a massive number of sensor devices interconnected, can be used for data exchange, intelligent identification, and management of interconnected “things.” IoT devices are proliferating and playing a crucial role in improving the living quality and living standard of the people. However, the real IoT is more vulnerable to attack by countless cyberattacks from the Internet, which may cause privacy data leakage, data tampering and also cause significant harm to society and individuals. Network security is essential in the IoT system, and Web injection is one of the most severe security problems, especially the webshell. To develop a safe IoT system, in this article, we apply essential machine learning models to detect webshell to build secure solutions for IoT network. Future, ensemble methods including random forest (RF), extremely randomized trees (ET), and Voting are used to improve the performances of these machine learning models. We also discuss webshell detection in lightweight and heavyweight computing scenarios for different IoT environments. Extensive experiments have been conducted on these models to verify the validity of webshell intrusion. Simulation results show that RF and ET are suitable for lightweight IoT scenarios, and Voting method is effective for heavyweight IoT scenarios.