Wrapper Feature Selection Research Articles

Predicting perinatal mortality based on maternal health status and health insurance service using homogeneous ensemble machine learning methods

BackgroundPerinatal mortality in Ethiopia is the highest in Africa, with 68 per 1000 pregnancies intrapartum deaths. It is mainly associated with home delivery, which contributes to more than 75% of perinatal deaths. Financial constraints significantly impact timely access to maternal health care. Financial incentives, such as health insurance, may address the demand- and supply-side factors. This study, hence, aims to predict perinatal mortality based on maternal health status and health insurance service using homogeneous ensemble machine learning methods.MethodsThe data was collected from the Ethiopian demographic health survey from 2011 to 2019 G.C. The data were pre-processed to get quality data that are suitable for the homogenous ensemble machine-learning algorithms to develop a model that predicts perinatal mortality. We have applied filter (chi-square and mutual information) and wrapper (sequential forward and sequential backward) feature selection methods. After selecting all the relevant features, we developed a predictive model using cat boost, random forest, and gradient boosting algorithms and evaluated the model using both objective (accuracy, precision, recall, F1_score, ROC) and subjective (domain expert) based evaluation techniques.ResultsPerinatal mortality prediction models were developed using random forest, gradient boosting, and cat boost algorithms with the overall accuracy of 89.95%, 90.24%, and 82%, respectively. Risk factors of perinatal mortality were identified using feature importance analysis and relevant rules were extracted using the best performing model.ConclusionsA prediction model that was developed using gradient boosting algorithms was selected for further use in the risk factor analysis, generating relevant rules, development of artifacts, and model deployment because it has registered better performance with 90.24% accuracy. The most determinant risk factors of perinatal mortality were identified using feature importance and some of them are community-based health insurance, mother's educational level, region and place of residence, age, wealth status, birth interval, preterm, smoking cigarette, anemia level, hemoglobin level, and marital status.

Tuberculous pleural effusion prediction using ant colony optimizer with grade-based search assisted support vector machine.

Although tuberculous pleural effusion (TBPE) is simply an inflammatory response of the pleura caused by tuberculosis infection, it can lead to pleural adhesions and cause sequelae of pleural thickening, which may severely affect the mobility of the chest cavity. In this study, we propose bGACO-SVM, a model with good diagnostic power, for the adjunctive diagnosis of TBPE. The model is based on an enhanced continuous ant colony optimization (ACOR) with grade-based search technique (GACO) and support vector machine (SVM) for wrapped feature selection. In GACO, grade-based search greatly improves the convergence performance of the algorithm and the ability to avoid getting trapped in local optimization, which improves the classification capability of bGACO-SVM. To test the performance of GACO, this work conducts comparative experiments between GACO and nine basic algorithms and nine state-of-the-art variants as well. Although the proposed GACO does not offer much advantage in terms of time complexity, the experimental results strongly demonstrate the core advantages of GACO. The accuracy of bGACO-predictive SVM was evaluated using existing datasets from the UCI and TBPE datasets. In the TBPE dataset trial, 147 TBPE patients were evaluated using the created bGACO-SVM model, showing that the bGACO-SVM method is an effective technique for accurately predicting TBPE.

Performance Analysis of Feature Subset Selection Techniques for Intrusion Detection

An intrusion detection system is one of the main defense lines used to provide security to data, information, and computer networks. The problems of this security system are the increased processing time, high false alarm rate, and low detection rate that occur due to the large amount of data containing various irrelevant and redundant features. Therefore, feature selection can solve this problem by reducing the number of features. Choosing appropriate feature selection methods that can reduce the number of features without a negative effect on the classification accuracy is a major challenge. This challenge motivated us to investigate the application of different wrapper feature selection techniques in intrusion detection. The performance of the selected techniques, such as the genetic algorithm (GA), sequential forward selection (SFS), and sequential backward selection (SBS), were analyzed, addressed, and compared to the existing techniques. The efficiency of the three feature selection techniques with two classification methods, including support vector machine (SVM) and multi perceptron (MLP), was compared. The CICIDS2017, CSE-CIC-IDS218, and NSL-KDD datasets were considered for the experiments. The efficiency of the proposed models was proved in the experimental results, which indicated that it had highest accuracy in the selected datasets.

Multi-surrogate assisted multi-objective evolutionary algorithms for feature selection in regression and classification problems with time series data

Feature selection wrapper methods are powerful mechanisms for reducing the complexity of prediction models while preserving and even improving their precision. Meta-heuristic methods, such as multi-objective evolutionary algorithms, are commonly used as search strategies in feature selection wrapper methods since they allow minimizing the cardinality of the attribute subset and simultaneously maximizing the predictive capacity of the model. However, in high-dimensional problems, multi-objective evolutionary algorithms for wrapper-type feature selection may require excessive computational time, sometimes impractical, especially when the learning algorithm has a high computational cost, such as deep learning. To address this drawback, in this paper we propose a multi-surrogate assisted multi-objective evolutionary algorithm for feature selection, specially designed to improve generalization error. The proposed method has been compared with conventional feature selection wrapper methods that use random forest, support vector machine and long short-term memory learning algorithms to evaluate subsets of attributes. The experiments have been carried out with regression and classification problems with time series data for air quality forecasting in the south-east of Spain and for indoor temperature forecasting in a domotic house. The results demonstrate the superiority of the proposed multi-surrogate assisted method over conventional wrapper methods using the same run times.

Whale social optimization driven deep recurrent neural network for loan eligibility prediction

SummaryPeople can use bank loans to make investment decisions thanks to technological advancements in the banking sector. The bank, however, only has a limited amount of resources, therefore it must grant to borrowers who can follow to the repayment schedule. Determining the candidate for a loan and finding a secure alternative for a bank are therefore crucial tasks. A deep recurrent neural network based on the whale social optimization algorithm (WSOA) is developed to predict loan eligibility. The WSOA is modified in this case to train Deep RNN for loan eligibility prediction in order to yield the most precise result. The proposed WSOA is developed by combining the social ski‐driver algorithm and the whale optimization algorithm. By selecting the trusted person to save bank assets, the risk factor is also decreased. The loan eligibility prediction is performed by processing massive data with Box‐Cox transformation. Moreover, wrapper feature selection is adapted for choosing imperative features. The proposed WSOA‐based Deep RNN provided increased efficiency with the highest accuracy of 94%, the highest sensitivity of 95.4%, and the highest specificity of 91.3%.

Implementation of Features Selection Based on Dragonfly Optimization Algorithm

Nowadays increasing dimensionality of data produces several issues in machine learning. Therefore, it is needed to decrease the number of features by choosing just the most important ones and eliminating duplicate features, also reducing the number of features that are important to the model. For this purpose, many methodologies known as Feature Selection are applied. In this study, a feature selection approach is proposed based on Swarm Intelligence methods, which search for the best points in the search area to achieve optimization. In this paper, a wrapper feature selection technique based on the Dragonfly algorithm is proposed. The dragonfly optimization technique is used to find the optimal subset of features that could accurately classify breast cancer as benign or malignant. Many times, the fitness function is defined as classification accuracy. In this study, hard vote classes are employed as a model developed to evaluate feature subsets that have been chosen. It is used as an evaluation function (fitness function) to evaluate each dragonfly in the population. The proposed ensemble hard voting classifier utilizes a combination of five machine-learning algorithms to produce a binary classification for feature selection: Support Vector Machine (SVM), K-Nearest Neighbors (K-NN), Naive Bayes (NB), Decision Tree (DT), and Random Forest (RF). According to the results of the experiments, the voting ensemble classifier has the greatest accuracy value among the single classifiers. The proposed method showed that when training the subset features, the accuracy generated by the voting classifier is high at 98.24%, whereas the training of all features achieved an accuracy of 96.49%. The proposed approach makes use of the UCI repository's Wisconsin Diagnostic Breast Cancer (WDBC) Dataset. Which consists of 569 instances and 30 features.

Chi2-MI: A hybrid feature selection based machine learning approach in diagnosis of chronic kidney disease

• Development of an intelligent diagnosis system to detect chronic kidney disease. • A hybrid wrapper feature selection method (Chi 2 -MI) has been proposed and applied. • Data pre-processing methods are adopted to prepare the dataset for the model. • Most impactful features based on correlation scores are selected to predict the CKD. • Extra tress classifier can diagnose CKD with 98% accuracy among 14 learning models. Early detection and characterization are considered crucial in treating and controlling the chronic renal disease. Because of the rising number of patients, the high risk of progression to end-stage renal disease, and the poor prognosis of morbidity and mortality, chronic kidney disease (CKD) is a significant burden on the healthcare system. Detecting CKD in its early stages is critical for saving millions of lives. The uniqueness of this study lies in developing a diagnosis system to detect chronic kidney disease using different Machine Learning (ML) algorithms with the support of a hybrid feature selection approach. This study exploited the 400 clinical data of CKD patients based on the dataset supplied by the University of California Irvine (UCI) available at their Machine Learning repository. Different data preparation techniques like encoding categorical features, missing values imputation, removing outlier factors, handling data imbalance, scaling data at the same level, and selecting relevant features are adopted to prepare the dataset for the prediction model. A hybrid Chi-squared test (Chi2) and Mutual Information (MI) based feature selection approach is proposed to remove redundant features, and a Pearson correlation matrix is also computed to consider the top important features for the prediction. Lastly, the Extra tress classifier can diagnose CKD with 98% accuracy and a 2% true negative rate without data leakage out of 14 machine learning models. On the other hand, the Bagging classifier performed worst with only 60% accuracy.

A two-stage hybrid gene selection algorithm combined with machine learning models to predict the rupture status in intracranial aneurysms.

An IA is an abnormal swelling of cerebral vessels, and a subset of these IAs can rupture causing aneurysmal subarachnoid hemorrhage (aSAH), often resulting in death or severe disability. Few studies have used an appropriate method of feature selection combined with machine learning by analyzing transcriptomic sequencing data to identify new molecular biomarkers. Following gene ontology (GO) and enrichment analysis, we found that the distinct status of IAs could lead to differential innate immune responses using all 913 differentially expressed genes, and considering that there are numerous irrelevant and redundant genes, we propose a mixed filter- and wrapper-based feature selection. First, we used the Fast Correlation-Based Filter (FCBF) algorithm to filter a large number of irrelevant and redundant genes in the raw dataset, and then used the wrapper feature selection method based on the he Multi-layer Perceptron (MLP) neural network and the Particle Swarm Optimization (PSO), accuracy (ACC) and mean square error (MSE) were then used as the evaluation criteria. Finally, we constructed a novel 10-gene signature (YIPF1, RAB32, WDR62, ANPEP, LRRCC1, AADAC, GZMK, WBP2NL, PBX1, and TOR1B) by the proposed two-stage hybrid algorithm FCBF-MLP-PSO and used different machine learning models to predict the rupture status in IAs. The highest ACC value increased from 0.817 to 0.919 (12.5% increase), the highest area under ROC curve (AUC) value increased from 0.87 to 0.94 (8.0% increase), and all evaluation metrics improved by approximately 10% after being processed by our proposed gene selection algorithm. Therefore, these 10 informative genes used to predict rupture status of IAs can be used as complements to imaging examinations in the clinic, meanwhile, this selected gene signature also provides new targets and approaches for the treatment of ruptured IAs.

A hierarchical machine learning model based on Glioblastoma patients' clinical, biomedical, and image data to analyze their treatment plans

Aim of studyGlioblastoma Multiforme (GBM) is an aggressive brain cancer in adults that kills most patients in the first year due to ineffective treatment. Different clinical, biomedical, and image data features are needed to analyze GBM, increasing complexities. Besides, they lead to weak performances for machine learning models due to ignoring physicians' knowledge. Therefore, this paper proposes a hierarchical model based on Fuzzy C-mean (FCM) clustering, Wrapper feature selection, and twelve classifiers to analyze treatment plans. Methodology/ApproachThe proposed method finds the effectiveness of previous and current treatment plans, hierarchically determining the best decision for future treatment plans for GBM patients using clinical data, biomedical data, and different image data. A case study is presented based on the Cancer Genome Atlas Glioblastoma Multiforme dataset to prove the effectiveness of the proposed model. This dataset is analyzed using data preprocessing, experts' knowledge, and a feature reduction method based on the Principal Component Analysis. Then, the FCM clustering method is utilized to reinforce classifier learning. Outcomes of studyThe proposed model finds the best combination of Wrapper feature selection and classifier for each cluster based on different measures, including accuracy, sensitivity, specificity, precision, F-score, and G-mean according to a hierarchical structure. It has the best performance among other reinforced classifiers. Besides, this model is compatible with real-world medical processes for GBM patients based on clinical, biomedical, and image data.

Motor Imagery BCI Classification Based on Multivariate Variational Mode Decomposition

In this article, a novel computer-aided diagnosis framework is proposed for the classification of motor imagery (MI) electroencephalogram (EEG) signals. First, a multivariate variational mode decomposition (MVMD) method was employed to obtain joint modes in frequency scale across all channels. Second, several multi-domain features (time domain, frequency domain, nonlinear and geometrical) were extracted from each EEG signal, and to further enhance the classification performance of different MI EEG signals, a variety of wrapper and filter feature selection methods were utilized with different channel combinations. Finally, to avoid a large number of training sessions for a particular device, extensive subject-independent experiments were performed. The MVMD applied to 18-channel EEG from the motor cortex area in combination with the ReliefF feature selection method achieved an average classification accuracy of 99.8% for a subject-dependent while 98.3% for subject-independent experiments. Besides the aforementioned combination provide above 99% accuracy for subjects with sufficient or small training samples for both subject-dependent or independent cases. These promising findings suggest that the proposed framework is flexible to use for subject-dependent or independent BCI systems.

Improving Air Pollution Prediction Modelling Using Wrapper Feature Selection

Feature selection is considered as one of the essential steps in data pre-processing. However, all of the previous studies on predicting PM10 concentration in Malaysia have been limited to statistical method feature selection, and none of these studies used machine-learning approaches. Therefore, the objective of this research is to investigate the influence variables of the PM10 prediction model by using wrapper feature selection to compare the prediction model performance of different wrapper feature selection and to predict the concentration of PM10 for the next day. This research uses 10 years of daily data on pollutant concentrations from two stations (Klang and Shah Alam) obtained from the Department of Environment Malaysia (DOE) from 2009 until 2018. Six wrapper methods (forward selection, backward elimination, stepwise, brute-force, weight-guided and genetic algorithm evolution and the predictive analytics multiple linear regression (MLR) and artificial neural network (ANN)) were implemented in this study. This study found that brute-force is the dominant wrapper method in most of the best models in selecting important features for MLR. Moreover, compared to MLR, ANN provides more advantages regarding model accuracy and permits feature selection in predicting PM10. The overall results revealed that the RMSE value for next day prediction in Klang is 20.728, while the AE value is 15.69. Furthermore, the RMSE value for next day prediction in Shah Alam is 10.004, while the AE value is 7.982. Finally, all of the predicted models in Klang and Shah Alam can be used to predict the PM10 concentrations. This proposed model can be used as a tool for an early warning system in giving air quality information to local authorities in order to formulate air-quality-improvement strategies.

Gömülü ve Sarıcı Öznitelik Seçim Yöntemleri Kullanılarak Akciğer Rahatsızlıklarının Tespiti

Son yıllarda biyomedikal sinyal işleme alanındaki gelişmelere rağmen, akciğer rahatsızlıklarının tespiti üzerine hızlı ve yüksek doğrulukta çalışan teşhis sistemlerine duyulan ihtiyaç artmaktadır. Yapılan çalışmada fiziki muayene ile 94 farklı kişiden, solunum döngülerinin otomatik olarak tespit edilmesiyle elde edilen 150 adet normal ve 444 adet normal olmayan akciğer sesleri veri tabanı olarak kullanılmıştır. Sınıflandırma işleminde öznitelik olarak frekans ve zaman bölgesinde 12 farklı yöntem uygulanmıştır. Tüm veriler %80 eğitim %20 test aşamasında kullanılacak şekilde ikiye bölünmüştür. Elde edilen öznitelikler gömülü ve sarıcı öznitelik seçim yöntemleri kullanılarak değerlendirilmiştir. Bu yöntemler; özyinelemeli öznitelik eliminasyonu, uyarlanabilir yapı öğrenimi ile öznitelik seçimi, bağımlılık kılavuzlu denetimsiz öznitelik seçimi, sıralı yerellik ile denetimsiz öznitelik seçimi, içbükey küçültme yoluyla öznitelik seçimi, en küçük mutlak büzülme ve seçim operatörü öznitelik seçim yöntemleri olarak isimlendirilmektedir. İncelenen bu öznitelikler doğrusal destek vektör makineleri, k en yakın komşuluk, karar ağaçları ve naive bayes yöntemleri ile sınıflandırılmıştır. Sonuç olarak öznitelik sayısının sınırlandırılmadığı durum için, özyinelemeli öznitelik eliminasyonu yönteminin k en yakın komşuluk sınıflandırma ile beraber kullanıldığı durum için %97,3 doğruluk değerindeki başarıma ulaşılmaktadır. Öznitelik sayısının üç ile sınırlandırıldığı durumda ise uyarlanabilir yapı öğrenimi ile öznitelik seçimi yönteminin karar ağaçları yöntemi ile beraber kullanılması ile %91,4 değerinde başarıma ulaşılmıştır.

Spatial mapping of landslide susceptibility in Jerash governorate of Jordan using genetic algorithm-based wrapper feature selection and bagging-based ensemble model

This study employs five genetic algorithm (GA)-based machine learning (ML) models, namely the Decision Tree (DT), k-Nearest Neighbors (kNN), NaïveBayes (NB), Support Vector Machine (SVM), and Extreme Learning Machine (ELM), to build a novel ensemble algorithm that is founded on the Bagging method for landslide susceptibility mapping (LSM) in Jerash Governorate, north of Jordan. The GA-based wrapper feature selection (FS) was done based on the five individual models and in the initial stages of modeling, an inquiry for the best feature for each of the five models was made. Finally, five hybrid models, namely DT-GA, kNN-GA, NB-GA, SVM-GA, and ELM-GA were constructed and combined to create Bagging-based ensemble model. The FS outcomes uncovered that rainfall depth, distance to roads, the Stream Power Index, the Normalized Difference Vegetation Index, slope, geology, and aspect are the most influential determinants of landslides. After the significant variables were identified, they were selected as input predictors and entered into the models. GA-based Bagging ensemble model with the area under the receiver operating characteristic curve (AUROC) of 0.85 achieved the highest accuracy in the validation run, followed by SVM-GA (AUROC = 0.80), NB-GA (AUROC = 0.76), DT-GA (AUROC = 0.72), kNN-GA (AUROC = 0.70), and ELM-GA (AUROC = 0.48).

Call Failure Prediction in IP Multimedia Subsystem (IMS) Networks

An explosion of traffic volume is the main driver behind launching various 5G services. The 5G network will utilize the IP Multimedia Subsystems (IMS) as a core network, same as in 4G networks. Thus, ensuring a high level of survivability and efficient failure management in the IMS is crucial before launching 5G services. We introduce a new methodology based on machine learning to predict the call failures occurring inside the IMS network using the traces for the Session Initiation Protocol (SIP) communication. Predicting that the call will fail enables the operator to prevent the failure by redirecting the call to another radio access technique by initiating the Circuit Switching fallback (CS-fallback) through a 380 SIP error response sent to the handset. The advantage of the model is not limited to call failure prediction, but also to know the root causes behind the failure; more specifically, the multi-factorial root is caused by using machine learning, which cannot be obtained using the traditional method (manual tracking of the traces). We built eight different machine learning models using four different classifiers (decision tree, naive Bayes, K-Nearest Neighbor (KNN), and Support Vector Machine (SVM)) and two different feature selection methods (Filter and Wrapper). Finally, we compare the different models and use the one with the highest prediction accuracy to obtain the root causes beyond the call failures. The results demonstrate that using SVM classifier with Wrapper feature selection method conducts the highest prediction accuracy, reaching 97.5%.

A new hybrid feature selection based on Improved Equilibrium Optimization

Feature selection is one of the most important preprocessing steps for analyzing high-dimensional data in the data mining and machine learning fields. In machine learning problems, Classification accuracy is highly dependent on the selected features of a dataset. The main purpose of feature selection is to remove additional and irrelevant features from the data feature set to increase the performance of algorithms. A novel feature selection model is proposed in this paper. In the first step, the features are scored based on the mRMR feature selection approach, and those with a higher score are selected. In the second step, the Wrapper feature selection approach is used to extract the best features based on the Improved Equilibrium Optimization (IMEO) algorithm. IMEO uses a new operator called Entropy-based to improve the performance of the Equilibrium Optimizer (EO) algorithm and prevents the algorithm from getting trap in local optima by controlling the exploration and exploitation abilities. Also, Levy flight is used to find new solutions in the search space and to improve the exploration phase of the EO algorithm. This mechanism will avoid local optimum by creating many short jumps and sometimes large jumps in the search space. The proposed IMEO algorithm is first tested on 23 benchmark test functions; after ensuring the efficiency of the proposed method, the Binary Improved Equilibrium Optimization (BIMEO) algorithm is defined based on the sigmoid transfer function to solve the feature selection problem. The simulation results of the BIMEO algorithm and state-of-the-art algorithms on 18 standard benchmark datasets show the performance of the proposed model compared to the other algorithms. In addition, the results of experiments on two chemical datasets prove the efficiency of the proposed algorithm, BIMEO, in terms of selective features and classification accuracy.

Wrapper feature selection method based differential evolution and extreme learning machine for intrusion detection system

The intrusion detection system (IDS) has gained a rapid increase of interest due to its widely recognized potential in various security fields, however, it suffers from several challenges. Different network datasets have several redundant and irrelevant features that affect the decision of the IDS classifier. Therefore, it is essential to decrease these features to improve the system performance. In this paper, an efficient wrapper feature selection method is proposed for improving the performance and decreasing the processing time of the IDS. The proposed approach employs a differential evaluation algorithm to select the useful features whilst the extreme learning machine classifier is applied after feature selection to evaluate the selected features. Many experiments are performed using the full NSL-KDD dataset to evaluate the performance of the proposed method. The results prove that the proposed approach can efficiently reduce the features, increase the accuracy, reduce the false alarm rates, and improve the processing time of the IDS in comparison to other recent related works.

Comparisons of Filter, Wrapper and Embedded-Based Feature Selection Techniques for Consistency of Software Metrics Analysis

Identifying and selecting the most consistent subset of metrics which improves the performance of software defect prediction model is paramount but challenging problem as it receives little attention in literature. The current research aimed at investigating the consistency of subsets of metrics that are produced by embedded feature selection techniques. Ten (10) feature selection techniques used from the families of filter and wrapper-based feature selection techniques commonly used in the defect prediction domain. Ten (10) publicly available defect datasets were studied which span both proprietary and open source domains. SVM-RFE-RF presented 42-93% consistent metrics across datasets. While the prior study on non-Embedded produced 56.5% consistent metrics at median. SVM-RFE-LF approach of Embedded Feature Selection Technique produced 54-80% consistent metrics across datasets and 42.5% at median. To state the purpose of tittle has been achieved Embedded based Feature Selection Techniques produced most efficient consistent subset selection across the entire datasets and amongst the feature selection techniques as compared with counterpart filter and wrapper-based feature selection techniques

Applications of dynamic feature selection and clustering methods to medical diagnosis

Machine learning methods are commonly used for disease and cancer diagnosis. The model performance can be improved via feature selection, feature reduction, and clustering methods. Although these supplementary techniques have certain advantages, they cannot necessarily guarantee better performance. The objective of this study is to improve the performance of classification methods used for medical diagnosis of various diseases. We propose a dynamic feature selection method based on the merits of both principal component analysis and Wrapper feature selection methods. It is a novel multi-objective feature selection method based on a customized genetic algorithm that is guided by eigenvalues of the features and feedbacks of various classifiers’ output. To reinforce classification learning and further enhancement of their performance, we utilize a dynamic selection of three clustering methods including K-means, fuzzy c-means, and particle swarm optimization. We also investigated the performance of two deep learning classifiers on the proposed methods.To show the impacts of combination of the proposed methods, we analyze the results of applying 12 machine learning and two deep learning classifiers to 30 imbalanced medical datasets. According to our extensive computational experiments and the statistical tests, the proposed dynamic feature selection and clustering methods perform significantly better than existing methods. The proposed methods not only improve the average of performance measures by 5% but also are more accurate than best performing classification methods in the literature used for the same datasets.

Predicting earnings management through machine learning ensemble classifiers

AbstractIn this paper, we utilize six novel ensemble classifiers to predict earnings management (EM) in both its forms, accrual‐based earnings management (AEM) and real earnings management (REM), and then compare the EM prediction accuracy of wrapper feature selection (FS) and filtering FS techniques in the context of EM. Specifically, we integrate three well‐known filtering FS techniques (information gain [IG], principal component analysis [PCA], and relief [Re]) and three popular wrapper FS techniques (particle swarm optimization [PSO], genetic algorithm [GA], and artificial bee colony [ABC]) with the support vector machine (SVM) to generate our ensemble classifiers. We then assess the performance of each of the six ensemble classifiers to predict AEM and REM based on three criteria: type Ι error, type ΙΙ error, and average accuracy. The results show that the ABC‐SVM ensemble classifier outperforms the others in predicting both AEM and REM. We also find that, overall, wrapper FS ensemble classifiers outperform filtering FS ensemble classifiers in predicting AEM and REM and that it is more difficult for our ensemble classifiers to predict REM than to predict AEM. This paper contributes to the literature on EM prediction by introducing six new ensemble classifiers. It is also the first work (to the best of our knowledge) in the domain of ensemble classifiers' applications (a) to consider both REM and AEM in one context and to show that REM is more difficult to predict than AEM and (b) to compare the performance of wrapper and filtering FS techniques in the EM prediction setting.

Suspended sediment load modeling using advanced hybrid rotation forest based elastic network approach

• New advanced machine learning (ML) models applied for suspended sediment load prediction. • Results compared with traditional ML models as a benchmark. • Manually and feature selection methods applied to explore the best input scenario. • Advanced hybrid Elastic Network model outperform of all developed models. • The outcomes have practical implications in watershed management applications. The distribution and transportation of suspended sediment load (Q ssl ) in rivers have a significant effect on the design of hydraulic structures, river morphology, water quality, and aquatic ecosystems. As direct measurement of Q ssl can be costly and time-consuming, reliable estimates are vital for watershed management. In the present study, four standalone models including an Elastic Network (EN), Alternating Model Tree (AM Tree), Reduced Error Pruning Tree (REP Tree) and the Dual Perturb and Combine Tree (DPC Tree), including four hybridized models that combine a standalone model with the Rotation Forest (RF), were developed and evaluated for Q ssl prediction at Talar Watershed, in the northern Iran. Multiple scenarios comprised of antecedent flow discharge (Q w ), rainfall (R) and Q ssl , with all constructed manually and automatically using a Wrapper Feature Selection (WFS) to predict the Q ssl at Shirgah hydrometric station from January 1, 2004 to September 22, 2019. The optimal model fitted results were evaluated using multiple graphical and quantitative metrics with the results revealing that the flow discharge is perhaps the best predictor of Q ssl . Based on the Nash-Sutcliffe Efficiency (NSE) metric, the RF-EN (NSE = 0.85), EN (NSE = 0.83), AM Tree (NSE = 0.79), RF-AM Tree (NSE = 0.81) and the RF-REP Tree (NSE = 0.79) models seemed to perform very well, with the REP Tree (NSE = 0.65) and RF-DPC (NSE = 0.71) performed well, while the DPC Tree model (NSE = 0.35) were notably unsatisfactory. The hybridized models, however, captured extreme values more accurately compared with the standalone models. Finally, the model outputs were compared to the well-known optimized ANFIS models with a metaheuristic approach (imperialist competitive algorithm (ICA) and BAT algorithms), and all these results revealed that most the of newly developed models outperformed the ANFIS-ICA and ANFIS-BAT algorithms. The new modelling approaches developed and testing using advanced hybrid Rotation Forest based Elastic Networks in this study have important practical implications for suspended sediment load modeling and applications.